🕷️ Crawler Inspector

URL Lookup

Direct Parameter Lookup

Raw Queries and Responses

1. Shard Calculation

Query:

Response:

Calculated Shard: 152 (from laksa105)

2. Crawled Status Check

Query:

curl -X POST \
  'http://laksa152.int.ahrefs:8124/' \
  -H 'Content-Type: text/plain' \
  -H 'X-ClickHouse-Database: crawler3' \
  -H 'Authorization: Basic YXBpOg==' \
  -d 'SELECT getAhrefsURLFromUnparsed(src_unparsed) AS found_url, ifNull(toUnixTimestamp(download_stamp), 0) AS crawl_time, ifNull(toUnixTimestamp(props_url_first_seen), 0) AS first_indexed_time, download_http_code AS http_code, src_unparsed AS src_unparsed, src_root_hash AS src_root_hash, history_drop_reason AS history_drop_reason, meta_title AS meta_title, meta_descriptions AS meta_descriptions, attrs_boilerpipe_text AS attrs_boilerpipe_text, attrs_markdown AS attrs_markdown, attrs_readable_markdown AS attrs_readable_markdown, meta_canonical AS meta_canonical FROM crawler3.page_info_local FINAL PREWHERE (src_root_hash, src_unparsed) IN ((getAhrefsRootHashFromUnparsed(getAhrefsUnparsedNoserviceFromURL(\'https://en.wikipedia.org/wiki/Natural_language_processing\')), getAhrefsUnparsedNoserviceFromURL(\'https://en.wikipedia.org/wiki/Natural_language_processing\'))) FORMAT JSONEachRow'

Response:

{"found_url":"https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing","crawl_time":1775808835,"first_indexed_time":1375978912,"http_code":200,"src_unparsed":"org,wikipedia!en,\/wiki\/Natural_language_processing s443","src_root_hash":"17790707453426894952","history_drop_reason":null,"meta_title":"Natural language processing - Wikipedia","meta_descriptions":[],"attrs_boilerpipe_text":"From Wikipedia, the free encyclopedia\nNatural language processing\n(\nNLP\n) is the processing of\nnatural language\ninformation by a\ncomputer\n. NLP is a subfield of\ncomputer science\nand is closely associated with\nartificial intelligence\n. NLP is also related to\ninformation retrieval\n,\nknowledge representation\n,\ncomputational linguistics\n, and\nlinguistics\nmore broadly.\n[\n1\n]\nMajor processing tasks in an NLP system include:\nspeech recognition\n,\ntext classification\n,\nnatural language understanding\n, and\nnatural language generation\n.\nNatural language processing has its roots in the 1950s.\n[\n2\n]\nAlready in 1950,\nAlan Turing\npublished an article titled \"\nComputing Machinery and Intelligence\n\" which proposed what is now called the\nTuring test\nas a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language.\nSymbolic NLP (1950s – early 1990s)\n[\nedit\n]\nA document parsed into an abstract syntax tree\nThe premise of symbolic NLP is often illustrated using\nJohn Searle's Chinese room\nthought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts.\n1950s\n: The\nGeorgetown experiment\nin 1954 involved fully\nautomatic translation\nof more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem.\n[\n3\n]\nHowever, real progress was much slower, and after the\nALPAC report\nin 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe\n[\n4\n]\n) until the late 1980s when the first\nstatistical machine translation\nsystems were developed.\n1960s\n: Some notably successful natural language processing systems developed in the 1960s were\nSHRDLU\n, a natural language system working in restricted \"\nblocks worlds\n\" with restricted vocabularies, and\nELIZA\n, a simulation of a\nRogerian psychotherapy\n, written by\nJoseph Weizenbaum\nbetween 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the \"patient\" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to \"My head hurts\" with \"Why do you say your head hurts?\". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only\ntwenty\nwords, because that was all that would fit in a computer  memory at the time.\n[\n5\n]\n1970s\n: During the 1970s, many programmers began to write \"conceptual\nontologies\n\", which structured real-world information into computer-understandable data.  Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981).  During this time, the first\nchatterbots\nwere written (e.g.,\nPARRY\n).\n1980s\n: The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of\nHPSG\nas a computational operationalization of\ngenerative grammar\n), morphology (e.g., two-level morphology\n[\n6\n]\n), semantics (e.g.,\nLesk algorithm\n), reference (e.g., within Centering Theory\n[\n7\n]\n) and other areas of natural language understanding (e.g., in the\nRhetorical Structure Theory\n). Other lines of research were continued, e.g., the development of chatterbots with\nRacter\nand\nJabberwacky\n. An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period.\n[\n8\n]\nStatistical NLP (1990s–present)\n[\nedit\n]\nUp until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of\nmachine learning\nalgorithms for language processing. This shift was influenced by increasing computational power (see\nMoore's law\n) and a decline in the dominance of\nChomskyan\nlinguistic theories... (e.g.\ntransformational grammar\n), whose theoretical underpinnings discouraged the sort of\ncorpus linguistics\nthat underlies the machine-learning approach to language processing.\n[\n9\n]\n1990s\n: Many of the notable early successes in statistical methods in NLP occurred in the field of\nmachine translation\n, due especially to work at IBM Research, such as\nIBM alignment models\n. These systems were able to take advantage of existing multilingual\ntextual corpora\nthat had been produced by the\nParliament of Canada\nand the\nEuropean Union\nas a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data.\n2000s\n: With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on\nunsupervised\nand\nsemi-supervised learning\nalgorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than\nsupervised learning\n, and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the\nWorld Wide Web\n), which can often make up for the worse efficiency if the algorithm used has a low enough\ntime complexity\nto be practical.\n2003:\nword n-gram model\n, at the time the best statistical algorithm, is outperformed by a\nmulti-layer perceptron\n(with a single hidden layer and context length of several words, trained on up to 14 million words, by\nBengio\net al.)\n[\n10\n]\n2010:\nTomáš Mikolov\n(then a PhD student at\nBrno University of Technology\n) with co-authors applied a simple\nrecurrent neural network\nwith a single hidden layer to language modeling,\n[\n11\n]\nand in the following years he went on to develop\nWord2vec\n. In the 2010s,\nrepresentation learning\nand\ndeep neural network\n-style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques\n[\n12\n]\n[\n13\n]\ncan achieve state-of-the-art results in many natural language tasks, e.g., in\nlanguage modeling\n[\n14\n]\nand parsing.\n[\n15\n]\n[\n16\n]\nThis is increasingly important\nin medicine and healthcare\n, where NLP helps analyze notes and text in\nelectronic health records\nthat would otherwise be inaccessible for study when seeking to improve care\n[\n17\n]\nor protect patient privacy.\n[\n18\n]\nApproaches: Symbolic, statistical, neural networks\n[\nedit\n]\nSymbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular:\n[\n19\n]\n[\n20\n]\nsuch as by writing grammars or devising heuristic rules for\nstemming\n.\nMachine learning\napproaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach:\nboth statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally.\nlanguage models\n, produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce.\nthe larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to\nintractability\nproblems.\nRule-based systems are commonly used:\nwhen the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the\nApertium\nsystem,\nfor preprocessing in NLP pipelines, e.g.,\ntokenization\n, or\nfor post-processing and transforming the output of NLP pipelines, e.g., for\nknowledge extraction\nfrom syntactic parses.\nStatistical approach\n[\nedit\n]\nIn the late 1980s and mid-1990s, the statistical approach ended a period of\nAI winter\n, which was caused by the inefficiencies of the rule-based approaches.\n[\n21\n]\n[\n22\n]\nThe earliest\ndecision trees\n, producing systems of hard\nif–then rules\n, were still very similar to the old rule-based approaches.\nOnly the introduction of hidden\nMarkov models\n, applied to part-of-speech tagging, announced the end of the old rule-based approach.\nA major drawback of statistical methods is that they require elaborate\nfeature engineering\n. Since 2015,\n[\n23\n]\nneural network\n–based methods have increasingly replaced traditional statistical approaches, using\nsemantic networks\n[\n24\n]\nand\nword embeddings\nto capture semantic properties of words.\nIntermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore.\nNeural machine translation\n, based on then-newly invented\nsequence-to-sequence\ntransformations, made obsolete the intermediate steps, such as word alignment, previously necessary for\nstatistical machine translation\n.\nThe following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks.\nThough natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below.\nText and speech processing\n[\nedit\n]\nWord cloud of stop words in Hebrew\nOptical character recognition\n(OCR)\nGiven an image representing printed text, determine the corresponding text.\nSpeech recognition\nGiven a sound clip of a person or people speaking, determine the textual representation of the speech.  This is the opposite of\ntext to speech\nand is one of the extremely difficult problems colloquially termed \"\nAI-complete\n\" (see above).  In\nnatural speech\nthere are hardly any pauses between successive words, and thus\nspeech segmentation\nis a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed\ncoarticulation\n, so the conversion of the\nanalog signal\nto discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent.\nSpeech segmentation\nGiven a sound clip of a person or people speaking, separate it into words.  A subtask of\nspeech recognition\nand typically grouped with it.\nText-to-speech\nGiven a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired.\n[\n25\n]\nWord segmentation\n(\nTokenization\n)\nTokenization\nis a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods.\n[\n26\n]\nFor a language like\nEnglish\n, this is fairly trivial, since words are usually separated by spaces. However, some written languages like\nChinese\n,\nJapanese\nand\nThai\ndo not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the\nvocabulary\nand\nmorphology\nof words in the language. Sometimes this process is also used in cases like\nbag of words\n(BOW) creation in data mining.\n[\n27\n]\n[\n28\n]\nMorphological analysis\n[\nedit\n]\nLemmatization of Basque words\nLemmatization\nThe task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form.\n[\n29\n]\nMorphological segmentation\nSeparate words into individual\nmorphemes\nand identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the\nmorphology\n(\ni.e.\n, the structure of words) of the language being considered.\nEnglish\nhas fairly simple morphology, especially\ninflectional morphology\n, and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., \"open, opens, opened, opening\") as separate words. In languages such as\nTurkish\nor\nMeitei\n, a highly\nagglutinated\nIndian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms.\n[\n30\n]\nPart-of-speech tagging\nGiven a sentence, determine the\npart of speech\n(POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, \"book\" can be a\nnoun\n(\"the book on the table\") or\nverb\n(\"to book a flight\"); \"set\" can be a noun, verb or\nadjective\n; and \"out\" can be any of at least five different parts of speech.\nStemming\nThe process of reducing inflected (or sometimes derived) words to a base form (e.g., \"close\" will be the root for \"closed\", \"closing\", \"close\", \"closer\" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary.\nGrammar induction\n[\n31\n]\nGenerate a\nformal grammar\nthat describes a language's syntax.\nSentence breaking\n(also known as \"\nsentence boundary disambiguation\n\")\nGiven a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by\nperiods\nor other\npunctuation marks\n, but these same characters can serve other purposes (e.g., marking\nabbreviations\n).\nParsing\nDetermine the\nparse tree\n(grammatical analysis) of a given sentence. The\ngrammar\nfor\nnatural languages\nis\nambiguous\nand typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing:\ndependency parsing\nand\nconstituency parsing\n. Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a\nprobabilistic context-free grammar\n(PCFG) (see also\nstochastic grammar\n).\nLexical semantics (of individual words in context)\n[\nedit\n]\nAn entity linking pipeline\nLexical semantics\nWhat is the computational meaning of individual words in context?\nDistributional semantics\nHow can we learn semantic representations from data?\nNamed entity recognition\n(NER)\nGiven a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although\ncapitalization\ncan aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of\nnamed entity\n, and in any case, is often inaccurate or insufficient.  For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized.  Furthermore, many other languages in non-Western scripts (e.g.\nChinese\nor\nArabic\n) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example,\nGerman\ncapitalizes all\nnouns\n, regardless of whether they are names, and\nFrench\nand\nSpanish\ndo not capitalize names that serve as\nadjectives\n. This task is also referred to as token classification.\n[\n32\n]\nSentiment analysis\n(see also\nMultimodal sentiment analysis\n)\nSentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as\nword n-grams\n,\nTerm Frequency-Inverse Document Frequency\n(TF-IDF) features, hand-generated features, or employ\ndeep learning\nmodels designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms.\n[\n26\n]\nTerminology extraction\nThe goal of terminology extraction is to automatically extract relevant terms from a given corpus.\nWord-sense disambiguation\n(WSD)\nMany words have more than one\nmeaning\n; we have to select the meaning which makes the most sense in context.  For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as\nWordNet\n.\nEntity linking\nMany words—typically proper names—refer to\nnamed entities\n; here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context.\nRelational semantics (semantics of individual sentences)\n[\nedit\n]\nRelationship extraction\nGiven a chunk of text, identify the relationships among named entities (e.g. who is married to whom).\nSemantic parsing\nGiven a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in\nAMR parsing\n) or in accordance with a logical formalism (e.g., in\nDRT parsing\n). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see\nNatural language understanding\nbelow).\nSemantic role labelling\n(see also implicit semantic role labelling below)\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal\nframes\n), then identify and classify the frame elements (\nsemantic roles\n).\nDiscourse (semantics beyond individual sentences)\n[\nedit\n]\nCoreference resolution\nGiven a sentence or larger chunk of text, determine which words (\"mentions\") refer to the same objects (\"entities\").\nAnaphora resolution\nis a specific example of this task, and is specifically concerned with matching up\npronouns\nwith the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called \"bridging relationships\" involving\nreferring expressions\n. For example, in a sentence such as \"He entered John's house through the front door\", \"the front door\" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to).\nDiscourse analysis\nThis rubric includes several related tasks.  One task is discourse parsing, i.e., identifying the\ndiscourse\nstructure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast).  Another possible task is recognizing and classifying the\nspeech acts\nin a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.).\nImplicit semantic role labelling\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal\nframes\n) and their explicit semantic roles in the current sentence (see\nSemantic role labelling\nabove). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to\npro-drop languages\n.\nRecognizing textual entailment\nGiven two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false.\n[\n33\n]\nTopic segmentation\nand recognition\nGiven a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment.\nArgument mining\nThe goal of argument mining is the automatic extraction and identification of argumentative structures from\nnatural language\ntext with the aid of computer programs.\n[\n34\n]\nSuch argumentative structures include the premise, conclusions, the\nargument scheme\nand the relationship between the main and subsidiary argument, or the main and counter-argument within discourse.\n[\n35\n]\n[\n36\n]\nHigher-level NLP applications\n[\nedit\n]\nMachine translation in Firefox\nAutomatic summarization\n(text summarization)\nProduce a readable summary of a chunk of text.  Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper.\nGrammatical error correction\nGrammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology\/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011.\n[\n37\n]\n[\n38\n]\n[\n39\n]\nAs far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as\nGPT-2\n, this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications.\nLogic translation\nTranslate a text from a natural language into formal logic.\nMachine translation\n(MT)\nAutomatically translate text from one human language to another.  This is one of the most difficult problems, and is a member of a class of problems colloquially termed \"\nAI-complete\n\", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly.\nNatural language understanding\n(NLU)\nConvert chunks of text into more formal representations such as\nfirst-order logic\nstructures that are easier for\ncomputer\nprograms to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as\nclosed-world assumption\n(CWA) vs.\nopen-world assumption\n, or subjective Yes\/No vs. objective True\/False is expected for the construction of a basis of semantics formalization.\n[\n40\n]\nNatural language generation\n(NLG):\nConvert information from computer databases or semantic intents into readable human language.\nBook generation\nNot an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter,\nThe policeman's beard is half-constructed\n).\n[\n41\n]\nThe first published work by a neural network was published in 2018,\n1 the Road\n, marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free)\nlanguage models\n. The first machine-generated science book was published in 2019 (Beta Writer,\nLithium-Ion Batteries\n, Springer, Cham).\n[\n42\n]\nUnlike\nRacter\nand\n1 the Road\n, this is grounded on factual knowledge and based on text summarization.\nDocument AI\nA Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants.\n[\n43\n]\nDialogue management\nComputer systems intended to converse with a human.\nQuestion answering\nGiven a human-language question, determine its answer. Typical questions have a specific right answer (such as \"What is the capital of Canada?\"), but sometimes open-ended questions are also considered (such as \"What is the meaning of life?\").\nText-to-image generation\nGiven a description of an image, generate an image that matches the description.\n[\n44\n]\nText-to-scene generation\nGiven a description of a scene, generate a\n3D model\nof the scene.\n[\n45\n]\n[\n46\n]\nText-to-video\nGiven a description of a video, generate a video that matches the description.\n[\n47\n]\n[\n48\n]\nGeneral tendencies and (possible) future directions\n[\nedit\n]\nBased on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed:\n[\n49\n]\nInterest on increasingly abstract, \"cognitive\" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09\/2017–18: dependency syntax, 2004–05\/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing).\nIncreasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+\/100+ languages)\nElimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems)\nMost higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above).\nCognition\nrefers to \"the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses.\"\n[\n50\n]\nCognitive science\nis the interdisciplinary, scientific study of the mind and its processes.\n[\n51\n]\nCognitive linguistics\nis an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics.\n[\n52\n]\nEspecially during the age of\nsymbolic NLP\n, the area of computational linguistics maintained strong ties with cognitive studies.\nAs an example,\nGeorge Lakoff\noffers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics,\n[\n53\n]\nwith two defining aspects:\nApply the theory of\nconceptual metaphor\n, explained by Lakoff as \"the understanding of one idea, in terms of another\" which provides an idea of the intent of the author.\n[\n54\n]\nFor example, consider the English word\nbig\n. When used in a comparison (\"That is a big tree\"), the author's intent is to imply that the tree is\nphysically large\nrelative to other trees or the authors experience.  When used metaphorically (\"Tomorrow is a big day\"), the author's intent to imply\nimportance\n.  The intent behind other usages, like in \"She is a big person\", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information.\nAssign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a\nprobabilistic context-free grammar\n(PCFG). The mathematical equation for such algorithms is presented in\nUS Patent 9269353\n:\n[\n55\n]\nWhere\nRMM\nis the relative measure of meaning\ntoken\nis any block of text, sentence, phrase or word\nN\nis the number of tokens being analyzed\nPMM\nis the probable measure of meaning based on a corpora\nd\nis the non zero location of the token along the sequence of\nN\ntokens\nPF\nis the probability function specific to a language\nTies with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar,\n[\n56\n]\nfunctional grammar,\n[\n57\n]\nconstruction grammar,\n[\n58\n]\ncomputational psycholinguistics and cognitive neuroscience (e.g.,\nACT-R\n), however, with limited uptake in mainstream NLP (as measured by presence on major conferences\n[\n59\n]\nof the\nACL\n). More recently, ideas of cognitive NLP have been revived as an approach to achieve\nexplainability\n, e.g., under the notion of \"cognitive AI\".\n[\n60\n]\nLikewise, ideas of cognitive NLP are inherent to neural models\nmultimodal\nNLP (although rarely made explicit)\n[\n61\n]\nand developments in\nartificial intelligence\n, specifically tools and technologies using\nlarge language model\napproaches\n[\n62\n]\nand new directions in\nartificial general intelligence\nbased on the\nfree energy principle\n[\n63\n]\nby British neuroscientist and theoretician at University College London\nKarl J. Friston\n.\n1 the Road\nArtificial intelligence detection software\nAutomated essay scoring\nBiomedical text mining\nCompound term processing\nComputational linguistics\nComputer-assisted reviewing\nControlled natural language\nDeep learning\nDeep linguistic processing\nDistributional semantics\nForeign language reading aid\nForeign language writing aid\nInformation extraction\nInformation retrieval\nLanguage and Communication Technologies\nLanguage model\nLanguage technology\nLatent semantic indexing\nMulti-agent system\nNative-language identification\nNatural-language programming\nNatural-language understanding\nNatural-language search\nOutline of natural language processing\nQuery expansion\nQuery understanding\nReification (linguistics)\nSpeech processing\nSpoken dialogue systems\nText-proofing\nText simplification\nTransformer (machine learning model)\nTruecasing\nQuestion answering\nWord2vec\n^\nEisenstein, Jacob (October 1, 2019).\nIntroduction to Natural Language Processing\n. The MIT Press. p. 1.\nISBN\n \n978-0-262-04284-0\n.\n^\n\"NLP\"\n.\n^\nHutchins, J. (2005).\n\"The history of machine translation in a nutshell\"\n(PDF)\n. Archived from\nthe original\n(PDF)\non 2019-07-13\n. Retrieved\n2019-02-04\n.\n[\nself-published source\n]\n^\n\"ALPAC: the (in)famous report\", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12.\n^\nCrevier 1993\n, pp. 146–148\n, see also\nBuchanan 2005\n, p. 56\n: \"Early programs were necessarily limited in scope by the size and speed of memory\"\n^\nKoskenniemi, Kimmo\n(1983),\nTwo-level morphology: A general computational model of word-form recognition and production\n(PDF)\n, Department of General Linguistics,\nUniversity of Helsinki\n, archived from\nthe original\n(PDF)\non 2018-12-21\n, retrieved\n2020-08-20\n^\nJoshi, A. K., & Weinstein, S. (1981, August).\nControl of Inference: Role of Some Aspects of Discourse Structure-Centering\n. In\nIJCAI\n(pp. 385–387).\n^\nGuida, G.; Mauri, G. (July 1986). \"Evaluation of natural language processing systems: Issues and approaches\".\nProceedings of the IEEE\n.\n74\n(7):\n1026–\n1035.\nBibcode\n:\n1986IEEEP..74.1026G\n.\ndoi\n:\n10.1109\/PROC.1986.13580\n.\nISSN\n \n1558-2256\n.\nS2CID\n \n30688575\n.\n^\nChomskyan linguistics encourages the investigation of \"\ncorner cases\n\" that stress the limits of its theoretical models (comparable to\npathological\nphenomena in mathematics), typically created using\nthought experiments\n, rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in\ncorpus linguistics\n.  The creation and use of such\ncorpora\nof real-world data is a fundamental part of machine-learning algorithms for natural language processing.  In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called \"\npoverty of the stimulus\n\" argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing.  As a result, the Chomskyan paradigm discouraged the application of such models to language processing.\n^\nBengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003).\n\"A neural probabilistic language model\"\n.\nThe Journal of Machine Learning Research\n.\n3\n:\n1137–\n1155 – via ACM Digital Library.\n^\nMikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010).\n\"Recurrent neural network based language model\"\n(PDF)\n.\nInterspeech 2010\n. pp. \n1045–\n1048.\ndoi\n:\n10.21437\/Interspeech.2010-343\n.\nS2CID\n \n17048224\n.\n^\nGoldberg, Yoav (2016).\n\"A Primer on Neural Network Models for Natural Language Processing\"\n.\nJournal of Artificial Intelligence Research\n.\n57\n:\n345–\n420.\narXiv\n:\n1807.10854\n.\ndoi\n:\n10.1613\/jair.4992\n.\nS2CID\n \n8273530\n.\n^\nGoodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016).\nDeep Learning\n. MIT Press.\n^\nJozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016).\nExploring the Limits of Language Modeling\n.\narXiv\n:\n1602.02410\n.\nBibcode\n:\n2016arXiv160202410J\n.\n^\nChoe, Do Kook; Charniak, Eugene.\n\"Parsing as Language Modeling\"\n.\nEmnlp 2016\n. Archived from\nthe original\non 2018-10-23\n. Retrieved\n2018-10-22\n.\n^\nVinyals, Oriol; et al. (2014).\n\"Grammar as a Foreign Language\"\n(PDF)\n.\nNips2015\n.\narXiv\n:\n1412.7449\n.\nBibcode\n:\n2014arXiv1412.7449V\n.\n^\nTurchin, Alexander; Florez Builes, Luisa F. (2021-03-19).\n\"Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review\"\n.\nJournal of Diabetes Science and Technology\n.\n15\n(3):\n553–\n560.\ndoi\n:\n10.1177\/19322968211000831\n.\nISSN\n \n1932-2968\n.\nPMC\n \n8120048\n.\nPMID\n \n33736486\n.\n^\nLee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10).\n\"Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study\"\n.\nJMIR Medical Informatics\n.\n10\n(6) e38482.\ndoi\n:\n10.2196\/38482\n.\nISSN\n \n2291-9694\n.\nPMC\n \n9233261\n.\nPMID\n \n35687381\n.\n^\nWinograd, Terry (1971).\nProcedures as a Representation for Data in a Computer Program for Understanding Natural Language\n(Thesis).\n^\nSchank, Roger C.; Abelson, Robert P. (1977).\nScripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures\n. Hillsdale: Erlbaum.\nISBN\n \n0-470-99033-3\n.\n^\nMark Johnson. How the statistical revolution changes (computational) linguistics.\nProceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics.\n^\nPhilip Resnik. Four revolutions.\nLanguage Log, February 5, 2011.\n^\nSocher, Richard.\n\"Deep Learning For NLP-ACL 2012 Tutorial\"\n.\nwww.socher.org\n. Archived from\nthe original\non 2021-04-14\n. Retrieved\n2020-08-17\n.\nThis was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. [Link is broken, try\nhttp:\/\/web.stanford.edu\/class\/cs224n\/\n]\n^\nSegev, Elad (2022).\nSemantic Network Analysis in Social Sciences\n. London: Routledge.\nISBN\n \n978-0-367-63652-4\n.\nArchived\nfrom the original on 5 December 2021\n. Retrieved\n5 December\n2021\n.\n^\nYi, Chucai;\nTian, Yingli\n(2012), \"Assistive Text Reading from Complex Background for Blind Persons\",\nCamera-Based Document Analysis and Recognition\n, Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. \n15–\n28,\nCiteSeerX\n \n10.1.1.668.869\n,\ndoi\n:\n10.1007\/978-3-642-29364-1_2\n,\nISBN\n \n978-3-642-29363-4\n{{\ncitation\n}}\n:  CS1 maint: work parameter with ISBN (\nlink\n)\n^\na\nb\n\"Natural Language Processing (NLP) - A Complete Guide\"\n.\nwww.deeplearning.ai\n. 2023-01-11\n. Retrieved\n2024-05-05\n.\n^\n\"GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks\"\n.\nGeeksforgeeks\n. 25 June 2024.\n^\nTorabi, Mohhamadreza; Ghasemi, Fahimeh.\n\"Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh\"\n.\nScientific Reports\n.\n1\n(1):\n1–\n15.\ndoi\n:\n10.1038\/s41598-025-14699-1\n.\nPMC\n \n12391535\n.\n^\n\"What is Natural Language Processing? Intro to NLP in Machine Learning\"\n.\nGyanSetu!\n. 2020-12-06\n. Retrieved\n2021-01-09\n.\n^\nKishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012).\n\"Manipuri Morpheme Identification\"\n(PDF)\n.\nProceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP)\n. COLING 2012, Mumbai, December 2012:\n95–\n108.\n{{\ncite journal\n}}\n:  CS1 maint: location (\nlink\n)\n^\nKlein, Dan; Manning, Christopher D. (2002).\n\"Natural language grammar induction using a constituent-context model\"\n(PDF)\n.\nAdvances in Neural Information Processing Systems\n.\n^\nKariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023).\n\"Precision information extraction for rare disease epidemiology at scale\"\n.\nJournal of Translational Medicine\n.\n21\n(1): 157.\ndoi\n:\n10.1186\/s12967-023-04011-y\n.\nPMC\n \n9972634\n.\nPMID\n \n36855134\n.\n^\nPASCAL Recognizing Textual Entailment Challenge (RTE-7)\nhttps:\/\/tac.nist.gov\/\/2011\/RTE\/\n^\nLippi, Marco; Torroni, Paolo (2016-04-20).\n\"Argumentation Mining: State of the Art and Emerging Trends\"\n.\nACM Transactions on Internet Technology\n.\n16\n(2):\n1–\n25.\ndoi\n:\n10.1145\/2850417\n.\nhdl\n:\n11585\/523460\n.\nISSN\n \n1533-5399\n.\nS2CID\n \n9561587\n.\n^\n\"Argument Mining – IJCAI2016 Tutorial\"\n.\nwww.i3s.unice.fr\n. Archived from\nthe original\non 2021-04-18\n. Retrieved\n2021-03-09\n.\n^\n\"NLP Approaches to Computational Argumentation – ACL 2016, Berlin\"\n. Retrieved\n2021-03-09\n.\n^\nAdministration.\n\"Centre for Language Technology (CLT)\"\n.\nMacquarie University\n. Retrieved\n2021-01-11\n.\n^\n\"Shared Task: Grammatical Error Correction\"\n.\nwww.comp.nus.edu.sg\n. Retrieved\n2021-01-11\n.\n^\n\"Shared Task: Grammatical Error Correction\"\n.\nwww.comp.nus.edu.sg\n. Retrieved\n2021-01-11\n.\n^\nDuan, Yucong; Cruz, Christophe (2011).\n\"Formalizing Semantic of Natural Language through Conceptualization from Existence\"\n.\nInternational Journal of Innovation, Management and Technology\n.\n2\n(1):\n37–\n42. Archived from\nthe original\non 2011-10-09.\n^\n\"U B U W E B :: Racter\"\n.\nwww.ubu.com\n. Retrieved\n2020-08-17\n.\n^\nWriter, Beta (2019).\nLithium-Ion Batteries\n.\ndoi\n:\n10.1007\/978-3-030-16800-1\n.\nISBN\n \n978-3-030-16799-8\n.\nS2CID\n \n155818532\n.\n^\n\"Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube\"\n.\nwww.youtube.com\n. 11 April 2019. Archived from\nthe original\non 2021-10-30\n. Retrieved\n2021-01-11\n.\n^\nRobertson, Adi (2022-04-06).\n\"OpenAI's DALL-E AI image generator can now edit pictures, too\"\n.\nThe Verge\n. Retrieved\n2022-06-07\n.\n^\n\"The Stanford Natural Language Processing Group\"\n.\nnlp.stanford.edu\n. Retrieved\n2022-06-07\n.\n^\nCoyne, Bob; Sproat, Richard (2001-08-01).\n\"WordsEye\"\n.\nProceedings of the 28th annual conference on Computer graphics and interactive techniques\n. SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. \n487–\n496.\ndoi\n:\n10.1145\/383259.383316\n.\nISBN\n \n978-1-58113-374-5\n.\nS2CID\n \n3842372\n.\n^\n\"Google announces AI advances in text-to-video, language translation, more\"\n.\nVentureBeat\n. 2022-11-02\n. Retrieved\n2022-11-09\n.\n^\nVincent, James (2022-09-29).\n\"Meta's new text-to-video AI generator is like DALL-E for video\"\n.\nThe Verge\n. Retrieved\n2022-11-09\n.\n^\n\"Previous shared tasks | CoNLL\"\n.\nwww.conll.org\n. Retrieved\n2021-01-11\n.\n^\n\"Cognition\"\n.\nLexico\n.\nOxford University Press\nand\nDictionary.com\n. Archived from\nthe original\non July 15, 2020\n. Retrieved\n6 May\n2020\n.\n^\n\"Ask the Cognitive Scientist\"\n.\nAmerican Federation of Teachers\n. 8 August 2014.\nCognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind.\n^\nRobinson, Peter (2008).\nHandbook of Cognitive Linguistics and Second Language Acquisition\n. Routledge. pp. \n3–\n8.\nISBN\n \n978-0-805-85352-0\n.\n^\nLakoff, George (1999).\nPhilosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm\n. New York Basic Books. pp. \n569–\n583.\nISBN\n \n978-0-465-05674-3\n.\n^\nStrauss, Claudia (1999).\nA Cognitive Theory of Cultural Meaning\n. Cambridge University Press. pp. \n156–\n164.\nISBN\n \n978-0-521-59541-4\n.\n^\nUS patent 9269353\n^\n\"Universal Conceptual Cognitive Annotation (UCCA)\"\n.\nUniversal Conceptual Cognitive Annotation (UCCA)\n. Retrieved\n2021-01-11\n.\n^\nRodríguez, F. C., & Mairal-Usón, R. (2016).\nBuilding an RRG computational grammar\n.\nOnomazein\n, (34), 86–117.\n^\n\"Fluid Construction Grammar – A fully operational processing system for construction grammars\"\n. Retrieved\n2021-01-11\n.\n^\n\"ACL Member Portal | The Association for Computational Linguistics Member Portal\"\n.\nwww.aclweb.org\n. Retrieved\n2021-01-11\n.\n^\n\"Chunks and Rules\"\n.\nW3C\n. Retrieved\n2021-01-11\n.\n^\nSocher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014).\n\"Grounded Compositional Semantics for Finding and Describing Images with Sentences\"\n.\nTransactions of the Association for Computational Linguistics\n.\n2\n:\n207–\n218.\ndoi\n:\n10.1162\/tacl_a_00177\n.\nS2CID\n \n2317858\n.\n^\nDasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). \"Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al\".\narXiv\n:\n2207.07051\n[\ncs.CL\n].\n^\nFriston, Karl J. (2022).\nActive Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference\n. The MIT Press.\nISBN\n \n978-0-262-36997-8\n.\nBates, M (1995).\n\"Models of natural language understanding\"\n.\nProceedings of the National Academy of Sciences of the United States of America\n.\n92\n(22):\n9977–\n9982.\nBibcode\n:\n1995PNAS...92.9977B\n.\ndoi\n:\n10.1073\/pnas.92.22.9977\n.\nPMC\n \n40721\n.\nPMID\n \n7479812\n.\nSteven Bird, Ewan Klein, and Edward Loper (2009).\nNatural Language Processing with Python\n. O'Reilly Media.\nISBN\n \n978-0-596-51649-9\n.\nKenna Hughes-Castleberry\n, \"A Murder Mystery Puzzle: The literary puzzle\nCain's Jawbone\n, which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms\",\nScientific American\n, vol. 329, no. 4 (November 2023), pp. 81–82. \"This murder mystery competition has revealed that although NLP (\nnatural-language processing\n) models are capable of incredible feats, their abilities are very much limited by the amount of\ncontext\nthey receive. This [...] could cause [difficulties] for researchers who hope to use them to do things such as analyze\nancient languages\n. In some cases, there are few historical records on long-gone\ncivilizations\nto serve as\ntraining data\nfor such a purpose.\" (p. 82.)\nDaniel Jurafsky and James H. Martin (2008).\nSpeech and Language Processing\n, 2nd edition. Pearson Prentice Hall.\nISBN\n \n978-0-13-187321-6\n.\nMohamed Zakaria Kurdi (2016).\nNatural Language Processing and Computational Linguistics: speech, morphology, and syntax\n, Volume 1. ISTE-Wiley.\nISBN\n \n978-1848218482\n.\nMohamed Zakaria Kurdi (2017).\nNatural Language Processing and Computational Linguistics: semantics, discourse, and applications\n, Volume 2. ISTE-Wiley.\nISBN\n \n978-1848219212\n.\nChristopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008).\nIntroduction to Information Retrieval\n. Cambridge University Press.\nISBN\n \n978-0-521-86571-5\n.\nOfficial html and pdf versions available without charge.\nChristopher D. Manning and Hinrich Schütze (1999).\nFoundations of Statistical Natural Language Processing\n. The MIT Press.\nISBN\n \n978-0-262-13360-9\n.\nDavid M. W. Powers and Christopher C. R. Turk (1989).\nMachine Learning of Natural Language\n. Springer-Verlag.\nISBN\n \n978-0-387-19557-5\n.\nMedia related to\nNatural language processing\nat Wikimedia Commons","attrs_markdown":"[Jump to content](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#bodyContent)\n\nMain menu\n\nMain menu\n\nmove to sidebar\n\nhide\n\nNavigation\n\n- [Main page](https:\/\/en.wikipedia.org\/wiki\/Main_Page \"Visit the main page [z]\")\n- [Contents](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Contents \"Guides to browsing Wikipedia\")\n- [Current events](https:\/\/en.wikipedia.org\/wiki\/Portal:Current_events \"Articles related to current events\")\n- [Random article](https:\/\/en.wikipedia.org\/wiki\/Special:Random \"Visit a randomly selected article [x]\")\n- [About Wikipedia](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:About \"Learn about Wikipedia and how it works\")\n- [Contact us](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Contact_us \"How to contact Wikipedia\")\n\nContribute\n\n- [Help](https:\/\/en.wikipedia.org\/wiki\/Help:Contents \"Guidance on how to use and edit Wikipedia\")\n- [Learn to edit](https:\/\/en.wikipedia.org\/wiki\/Help:Introduction \"Learn how to edit Wikipedia\")\n- [Community portal](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Community_portal \"The hub for editors\")\n- [Recent changes](https:\/\/en.wikipedia.org\/wiki\/Special:RecentChanges \"A list of recent changes to Wikipedia [r]\")\n- [Upload file](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:File_upload_wizard \"Add images or other media for use on Wikipedia\")\n- [Special pages](https:\/\/en.wikipedia.org\/wiki\/Special:SpecialPages \"A list of all special pages [q]\")\n\n[![](https:\/\/en.wikipedia.org\/static\/images\/icons\/enwiki-25.svg) ![Wikipedia](https:\/\/en.wikipedia.org\/static\/images\/mobile\/copyright\/wikipedia-wordmark-en-25.svg) ![The Free Encyclopedia](https:\/\/en.wikipedia.org\/static\/images\/mobile\/copyright\/wikipedia-tagline-en-25.svg)](https:\/\/en.wikipedia.org\/wiki\/Main_Page)\n\n[Search](https:\/\/en.wikipedia.org\/wiki\/Special:Search \"Search Wikipedia [f]\")\n\nAppearance\n\n- [Donate](https:\/\/donate.wikimedia.org\/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en)\n- [Create account](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:CreateAccount&returnto=Natural+language+processing \"You are encouraged to create an account and log in; however, it is not mandatory\")\n- [Log in](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:UserLogin&returnto=Natural+language+processing \"You're encouraged to log in; however, it's not mandatory. [o]\")\n\nPersonal tools\n\n- [Donate](https:\/\/donate.wikimedia.org\/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en)\n- [Create account](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:CreateAccount&returnto=Natural+language+processing \"You are encouraged to create an account and log in; however, it is not mandatory\")\n- [Log in](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:UserLogin&returnto=Natural+language+processing \"You're encouraged to log in; however, it's not mandatory. [o]\")\n\n## Contents\nmove to sidebar\n\nhide\n\n- [(Top)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing)\n- [1 History](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#History)\n  Toggle History subsection\n  - [1\\.1 Symbolic NLP (1950s – early 1990s)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Symbolic_NLP_\\(1950s_%E2%80%93_early_1990s\\))\n  - [1\\.2 Statistical NLP (1990s–present)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Statistical_NLP_\\(1990s%E2%80%93present\\))\n- [2 Approaches: Symbolic, statistical, neural networks](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Approaches:_Symbolic,_statistical,_neural_networks)\n  Toggle Approaches: Symbolic, statistical, neural networks subsection\n  - [2\\.1 Statistical approach](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Statistical_approach)\n  - [2\\.2 Neural networks](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Neural_networks)\n- [3 Common NLP tasks](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Common_NLP_tasks)\n  Toggle Common NLP tasks subsection\n  - [3\\.1 Text and speech processing](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Text_and_speech_processing)\n  - [3\\.2 Morphological analysis](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Morphological_analysis)\n  - [3\\.3 Syntactic analysis](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Syntactic_analysis)\n  - [3\\.4 Lexical semantics (of individual words in context)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Lexical_semantics_\\(of_individual_words_in_context\\))\n  - [3\\.5 Relational semantics (semantics of individual sentences)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Relational_semantics_\\(semantics_of_individual_sentences\\))\n  - [3\\.6 Discourse (semantics beyond individual sentences)](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Discourse_\\(semantics_beyond_individual_sentences\\))\n  - [3\\.7 Higher-level NLP applications](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Higher-level_NLP_applications)\n- [4 General tendencies and (possible) future directions](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#General_tendencies_and_\\(possible\\)_future_directions)\n  Toggle General tendencies and (possible) future directions subsection\n  - [4\\.1 Cognition](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Cognition)\n- [5 See also](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#See_also)\n- [6 References](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#References)\n- [7 Further reading](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Further_reading)\n- [8 External links](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#External_links)\n\nToggle the table of contents\n\n# Natural language processing\n71 languages\n\n- [Afrikaans](https:\/\/af.wikipedia.org\/wiki\/Natuurliketaalverwerking \"Natuurliketaalverwerking – Afrikaans\")\n- [العربية](https:\/\/ar.wikipedia.org\/wiki\/%D9%85%D8%B9%D8%A7%D9%84%D8%AC%D8%A9_%D8%A7%D9%84%D9%84%D8%BA%D8%A9_%D8%A7%D9%84%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A%D8%A9 \"معالجة اللغة الطبيعية – Arabic\")\n- [مصرى](https:\/\/arz.wikipedia.org\/wiki\/%D8%AA%D8%AD%D9%84%D9%8A%D9%84_%D8%A7%D9%84%D9%84%D8%BA%D8%A7%D8%AA_%D8%A7%D9%84%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A%D9%87 \"تحليل اللغات الطبيعيه – Egyptian Arabic\")\n- [Azərbaycanca](https:\/\/az.wikipedia.org\/wiki\/T%C9%99bii_dilin_emal%C4%B1 \"Təbii dilin emalı – Azerbaijani\")\n- [Беларуская (тарашкевіца)](https:\/\/be-tarask.wikipedia.org\/wiki\/%D0%90%D0%BF%D1%80%D0%B0%D1%86%D0%BE%D1%9E%D0%BA%D0%B0_%D0%BD%D0%B0%D1%82%D1%83%D1%80%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D0%B9_%D0%BC%D0%BE%D0%B2%D1%8B \"Апрацоўка натуральнай мовы – Belarusian (Taraškievica orthography)\")\n- [Беларуская](https:\/\/be.wikipedia.org\/wiki\/%D0%90%D0%BF%D1%80%D0%B0%D1%86%D0%BE%D1%9E%D0%BA%D0%B0_%D0%BD%D0%B0%D1%82%D1%83%D1%80%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D0%B9_%D0%BC%D0%BE%D0%B2%D1%8B \"Апрацоўка натуральнай мовы – Belarusian\")\n- [Български](https:\/\/bg.wikipedia.org\/wiki\/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%BD%D0%B0_%D0%B5%D1%81%D1%82%D0%B5%D1%81%D1%82%D0%B2%D0%B5%D0%BD_%D0%B5%D0%B7%D0%B8%D0%BA \"Обработка на естествен език – Bulgarian\")\n- [বাংলা](https:\/\/bn.wikipedia.org\/wiki\/%E0%A6%B8%E0%A7%8D%E0%A6%AC%E0%A6%BE%E0%A6%AD%E0%A6%BE%E0%A6%AC%E0%A6%BF%E0%A6%95_%E0%A6%AD%E0%A6%BE%E0%A6%B7%E0%A6%BE_%E0%A6%AA%E0%A7%8D%E0%A6%B0%E0%A6%95%E0%A7%8D%E0%A6%B0%E0%A6%BF%E0%A6%AF%E0%A6%BC%E0%A6%BE%E0%A6%9C%E0%A6%BE%E0%A6%A4%E0%A6%95%E0%A6%B0%E0%A6%A3 \"স্বাভাবিক ভাষা প্রক্রিয়াজাতকরণ – Bangla\")\n- [Brezhoneg](https:\/\/br.wikipedia.org\/wiki\/Treterezh_emgefre_al_lavar \"Treterezh emgefre al lavar – Breton\")\n- [Bosanski](https:\/\/bs.wikipedia.org\/wiki\/Obrada_prirodnog_jezika \"Obrada prirodnog jezika – Bosnian\")\n- [Català](https:\/\/ca.wikipedia.org\/wiki\/Processament_del_llenguatge_natural \"Processament del llenguatge natural – Catalan\")\n- [کوردی](https:\/\/ckb.wikipedia.org\/wiki\/%D9%BE%DB%8E%D9%88%D8%A7%DA%98%DB%86%DA%A9%D8%B1%D8%AF%D9%86%DB%8C_%D8%B2%D9%85%D8%A7%D9%86%DB%8C_%D8%B3%D8%B1%D9%88%D8%B4%D8%AA%DB%8C \"پێواژۆکردنی زمانی سروشتی – Central Kurdish\")\n- [Čeština](https:\/\/cs.wikipedia.org\/wiki\/Zpracov%C3%A1n%C3%AD_p%C5%99irozen%C3%A9ho_jazyka \"Zpracování přirozeného jazyka – Czech\")\n- [Cymraeg](https:\/\/cy.wikipedia.org\/wiki\/Prosesu_Iaith_Naturiol \"Prosesu Iaith Naturiol – Welsh\")\n- [Dansk](https:\/\/da.wikipedia.org\/wiki\/Sprogteknologi \"Sprogteknologi – Danish\")\n- [Deutsch](https:\/\/de.wikipedia.org\/wiki\/Verarbeitung_nat%C3%BCrlicher_Sprache \"Verarbeitung natürlicher Sprache – German\")\n- [Ελληνικά](https:\/\/el.wikipedia.org\/wiki\/%CE%95%CF%80%CE%B5%CE%BE%CE%B5%CF%81%CE%B3%CE%B1%CF%83%CE%AF%CE%B1_%CF%86%CF%85%CF%83%CE%B9%CE%BA%CE%AE%CF%82_%CE%B3%CE%BB%CF%8E%CF%83%CF%83%CE%B1%CF%82 \"Επεξεργασία φυσικής γλώσσας – Greek\")\n- [Esperanto](https:\/\/eo.wikipedia.org\/wiki\/Natur-lingva_prilaborado \"Natur-lingva prilaborado – Esperanto\")\n- [Español](https:\/\/es.wikipedia.org\/wiki\/Procesamiento_de_lenguajes_naturales \"Procesamiento de lenguajes naturales – Spanish\")\n- [Eesti](https:\/\/et.wikipedia.org\/wiki\/Loomuliku_keele_t%C3%B6%C3%B6tlus \"Loomuliku keele töötlus – Estonian\")\n- [Euskara](https:\/\/eu.wikipedia.org\/wiki\/Hizkuntzaren_prozesamendu \"Hizkuntzaren prozesamendu – Basque\")\n- [فارسی](https:\/\/fa.wikipedia.org\/wiki\/%D9%BE%D8%B1%D8%AF%D8%A7%D8%B2%D8%B4_%D8%B2%D8%A8%D8%A7%D9%86%E2%80%8C%D9%87%D8%A7%DB%8C_%D8%B7%D8%A8%DB%8C%D8%B9%DB%8C \"پردازش زبان‌های طبیعی – Persian\")\n- [Suomi](https:\/\/fi.wikipedia.org\/wiki\/Luonnollisen_kielen_k%C3%A4sittely \"Luonnollisen kielen käsittely – Finnish\")\n- [Français](https:\/\/fr.wikipedia.org\/wiki\/Traitement_automatique_des_langues \"Traitement automatique des langues – French\")\n- [Gaeilge](https:\/\/ga.wikipedia.org\/wiki\/Pr%C3%B3ise%C3%A1il_teanga_n%C3%A1d%C3%BArtha \"Próiseáil teanga nádúrtha – Irish\")\n- [Galego](https:\/\/gl.wikipedia.org\/wiki\/Procesamento_da_linguaxe_natural \"Procesamento da linguaxe natural – Galician\")\n- [עברית](https:\/\/he.wikipedia.org\/wiki\/%D7%A2%D7%99%D7%91%D7%95%D7%93_%D7%A9%D7%A4%D7%94_%D7%98%D7%91%D7%A2%D7%99%D7%AA \"עיבוד שפה טבעית – Hebrew\")\n- [हिन्दी](https:\/\/hi.wikipedia.org\/wiki\/%E0%A4%AA%E0%A5%8D%E0%A4%B0%E0%A4%BE%E0%A4%95%E0%A5%83%E0%A4%A4%E0%A4%BF%E0%A4%95_%E0%A4%AD%E0%A4%BE%E0%A4%B7%E0%A4%BE_%E0%A4%B8%E0%A4%82%E0%A4%B8%E0%A4%BE%E0%A4%A7%E0%A4%A8 \"प्राकृतिक भाषा संसाधन – Hindi\")\n- [Hrvatski](https:\/\/hr.wikipedia.org\/wiki\/Obrada_prirodnog_jezika \"Obrada prirodnog jezika – Croatian\")\n- [Հայերեն](https:\/\/hy.wikipedia.org\/wiki\/%D4%B2%D5%B6%D5%A1%D5%AF%D5%A1%D5%B6_%D5%AC%D5%A5%D5%A6%D5%BE%D5%AB_%D5%B4%D5%B7%D5%A1%D5%AF%D5%B8%D6%82%D5%B4 \"Բնական լեզվի մշակում – Armenian\")\n- [Արեւմտահայերէն](https:\/\/hyw.wikipedia.org\/wiki\/%D4%B2%D5%B6%D5%A1%D5%AF%D5%A1%D5%B6_%D4%BC%D5%A5%D5%A6%D5%B8%D6%82%D5%AB_%D5%84%D5%B7%D5%A1%D5%AF%D5%B8%D6%82%D5%B4 \"Բնական Լեզուի Մշակում – Western Armenian\")\n- [Bahasa Indonesia](https:\/\/id.wikipedia.org\/wiki\/Pengolahan_bahasa_alami \"Pengolahan bahasa alami – Indonesian\")\n- [Ido](https:\/\/io.wikipedia.org\/wiki\/Procedo_pri_naturala_linguo \"Procedo pri naturala linguo – Ido\")\n- [Íslenska](https:\/\/is.wikipedia.org\/wiki\/M%C3%A1lgreining \"Málgreining – Icelandic\")\n- [Italiano](https:\/\/it.wikipedia.org\/wiki\/Elaborazione_del_linguaggio_naturale \"Elaborazione del linguaggio naturale – Italian\")\n- [日本語](https:\/\/ja.wikipedia.org\/wiki\/%E8%87%AA%E7%84%B6%E8%A8%80%E8%AA%9E%E5%87%A6%E7%90%86 \"自然言語処理 – Japanese\")\n- [ქართული](https:\/\/ka.wikipedia.org\/wiki\/%E1%83%91%E1%83%A3%E1%83%9C%E1%83%94%E1%83%91%E1%83%A0%E1%83%98%E1%83%95%E1%83%98_%E1%83%94%E1%83%9C%E1%83%98%E1%83%A1_%E1%83%93%E1%83%90%E1%83%9B%E1%83%A3%E1%83%A8%E1%83%90%E1%83%95%E1%83%94%E1%83%91%E1%83%90 \"ბუნებრივი ენის დამუშავება – Georgian\")\n- [Qaraqalpaqsha](https:\/\/kaa.wikipedia.org\/wiki\/T%C3%A1biyiy_tildi_qayta_islew \"Tábiyiy tildi qayta islew – Kara-Kalpak\")\n- [ಕನ್ನಡ](https:\/\/kn.wikipedia.org\/wiki\/%E0%B2%B8%E0%B2%B9%E0%B2%9C_%E0%B2%AD%E0%B2%BE%E0%B2%B7%E0%B2%BE_%E0%B2%B8%E0%B2%82%E0%B2%B8%E0%B3%8D%E0%B2%95%E0%B2%B0%E0%B2%A3%E0%B3%86 \"ಸಹಜ ಭಾಷಾ ಸಂಸ್ಕರಣೆ – Kannada\")\n- [한국어](https:\/\/ko.wikipedia.org\/wiki\/%EC%9E%90%EC%97%B0%EC%96%B4_%EC%B2%98%EB%A6%AC \"자연어 처리 – Korean\")\n- [Lietuvių](https:\/\/lt.wikipedia.org\/wiki\/Nat%C5%ABraliosios_kalbos_apdorojimas \"Natūraliosios kalbos apdorojimas – Lithuanian\")\n- [Latviešu](https:\/\/lv.wikipedia.org\/wiki\/Dabisk%C4%81s_valodas_apstr%C4%81de \"Dabiskās valodas apstrāde – Latvian\")\n- [Македонски](https:\/\/mk.wikipedia.org\/wiki\/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%BD%D0%B0_%D0%BF%D1%80%D0%B8%D1%80%D0%BE%D0%B4%D0%BD%D0%B8_%D1%98%D0%B0%D0%B7%D0%B8%D1%86%D0%B8 \"Обработка на природни јазици – Macedonian\")\n- [Монгол](https:\/\/mn.wikipedia.org\/wiki\/%D0%9A%D0%BE%D0%BC%D0%BF%D1%8C%D1%8E%D1%82%D0%B5%D1%80%D1%8B%D0%BD_%D1%85%D1%8D%D0%BB_%D1%88%D0%B8%D0%BD%D0%B6%D0%BB%D1%8D%D0%BB \"Компьютерын хэл шинжлэл – Mongolian\")\n- [मराठी](https:\/\/mr.wikipedia.org\/wiki\/%E0%A4%A8%E0%A5%88%E0%A4%B8%E0%A4%B0%E0%A5%8D%E0%A4%97%E0%A4%BF%E0%A4%95_%E0%A4%AD%E0%A4%BE%E0%A4%B7%E0%A4%BE_%E0%A4%AA%E0%A5%8D%E0%A4%B0%E0%A4%95%E0%A5%8D%E0%A4%B0%E0%A4%BF%E0%A4%AF%E0%A4%BE \"नैसर्गिक भाषा प्रक्रिया – Marathi\")\n- [မြန်မာဘာသာ](https:\/\/my.wikipedia.org\/wiki\/%E1%80%99%E1%80%AD%E1%80%81%E1%80%84%E1%80%BA%E1%80%98%E1%80%AC%E1%80%9E%E1%80%AC%E1%80%85%E1%80%80%E1%80%AC%E1%80%B8%E1%80%9E%E1%80%AF%E1%80%B6%E1%80%B8_%E1%80%80%E1%80%BD%E1%80%94%E1%80%BA%E1%80%95%E1%80%BB%E1%80%B0%E1%80%90%E1%80%AC%E1%80%85%E1%80%94%E1%80%85%E1%80%BA \"မိခင်ဘာသာစကားသုံး ကွန်ပျူတာစနစ် – Burmese\")\n- [Nederlands](https:\/\/nl.wikipedia.org\/wiki\/Taaltechnologie \"Taaltechnologie – Dutch\")\n- [Norsk bokmål](https:\/\/no.wikipedia.org\/wiki\/Spr%C3%A5kteknologi \"Språkteknologi – Norwegian Bokmål\")\n- [ଓଡ଼ିଆ](https:\/\/or.wikipedia.org\/wiki\/%E0%AC%A8%E0%AD%8D%E0%AD%9F%E0%AC%BE%E0%AC%9A%E0%AD%81%E0%AC%B0%E0%AC%BE%E0%AC%B2_%E0%AC%B2%E0%AC%BE%E0%AC%99%E0%AD%8D%E0%AC%97%E0%AD%81%E0%AC%8F%E0%AC%9C_%E0%AC%AA%E0%AD%8D%E0%AC%B0%E0%AD%8B%E0%AC%B8%E0%AD%87%E0%AC%B8%E0%AC%BF%E0%AC%82 \"ନ୍ୟାଚୁରାଲ ଲାଙ୍ଗୁଏଜ ପ୍ରୋସେସିଂ – Odia\")\n- [Picard](https:\/\/pcd.wikipedia.org\/wiki\/Traitemint_automatique_d%27ches_langues \"Traitemint automatique d'ches langues – Picard\")\n- [Polski](https:\/\/pl.wikipedia.org\/wiki\/Przetwarzanie_j%C4%99zyka_naturalnego \"Przetwarzanie języka naturalnego – Polish\")\n- [Piemontèis](https:\/\/pms.wikipedia.org\/wiki\/NLP \"NLP – Piedmontese\")\n- [پښتو](https:\/\/ps.wikipedia.org\/wiki\/%D8%AF_%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A_%DA%98%D8%A8%DB%90_%D9%BE%D8%B1%D9%88%D8%B3%D8%B3_%DA%A9%D9%88%D9%84 \"د طبيعي ژبې پروسس کول – Pashto\")\n- [Português](https:\/\/pt.wikipedia.org\/wiki\/Processamento_de_linguagem_natural \"Processamento de linguagem natural – Portuguese\")\n- [Runa Simi](https:\/\/qu.wikipedia.org\/wiki\/Paqariq_simi_thatkichay \"Paqariq simi thatkichay – Quechua\")\n- [Română](https:\/\/ro.wikipedia.org\/wiki\/Prelucrarea_limbajului_natural \"Prelucrarea limbajului natural – Romanian\")\n- [Русский](https:\/\/ru.wikipedia.org\/wiki\/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%B5%D1%81%D1%82%D0%B5%D1%81%D1%82%D0%B2%D0%B5%D0%BD%D0%BD%D0%BE%D0%B3%D0%BE_%D1%8F%D0%B7%D1%8B%D0%BA%D0%B0 \"Обработка естественного языка – Russian\")\n- [Srpskohrvatski \/ српскохрватски](https:\/\/sh.wikipedia.org\/wiki\/Obrada_prirodnih_jezika \"Obrada prirodnih jezika – Serbo-Croatian\")\n- [Simple English](https:\/\/simple.wikipedia.org\/wiki\/Natural_language_processing \"Natural language processing – Simple English\")\n- [Shqip](https:\/\/sq.wikipedia.org\/wiki\/P%C3%ABrpunimi_i_gjuh%C3%ABs_natyrore \"Përpunimi i gjuhës natyrore – Albanian\")\n- [Српски \/ srpski](https:\/\/sr.wikipedia.org\/wiki\/Obrada_prirodnih_jezika \"Obrada prirodnih jezika – Serbian\")\n- [தமிழ்](https:\/\/ta.wikipedia.org\/wiki\/%E0%AE%87%E0%AE%AF%E0%AE%B1%E0%AF%8D%E0%AE%95%E0%AF%88_%E0%AE%AE%E0%AF%8A%E0%AE%B4%E0%AE%BF_%E0%AE%AE%E0%AF%81%E0%AE%B1%E0%AF%88%E0%AE%AF%E0%AE%BE%E0%AE%95%E0%AF%8D%E0%AE%95%E0%AE%AE%E0%AF%8D \"இயற்கை மொழி முறையாக்கம் – Tamil\")\n- [తెలుగు](https:\/\/te.wikipedia.org\/wiki\/%E0%B0%B8%E0%B0%B9%E0%B0%9C_%E0%B0%AD%E0%B0%BE%E0%B0%B7%E0%B0%BE_%E0%B0%AA%E0%B1%8D%E0%B0%B0%E0%B0%95%E0%B1%8D%E0%B0%B0%E0%B0%BF%E0%B0%AF \"సహజ భాషా ప్రక్రియ – Telugu\")\n- [ไทย](https:\/\/th.wikipedia.org\/wiki\/%E0%B8%81%E0%B8%B2%E0%B8%A3%E0%B8%9B%E0%B8%A3%E0%B8%B0%E0%B8%A1%E0%B8%A7%E0%B8%A5%E0%B8%A0%E0%B8%B2%E0%B8%A9%E0%B8%B2%E0%B8%98%E0%B8%A3%E0%B8%A3%E0%B8%A1%E0%B8%8A%E0%B8%B2%E0%B8%95%E0%B8%B4 \"การประมวลภาษาธรรมชาติ – Thai\")\n- [Türkçe](https:\/\/tr.wikipedia.org\/wiki\/Do%C4%9Fal_dil_i%C5%9Fleme \"Doğal dil işleme – Turkish\")\n- [Українська](https:\/\/uk.wikipedia.org\/wiki\/%D0%9E%D0%B1%D1%80%D0%BE%D0%B1%D0%BA%D0%B0_%D0%BF%D1%80%D0%B8%D1%80%D0%BE%D0%B4%D0%BD%D0%BE%D1%97_%D0%BC%D0%BE%D0%B2%D0%B8 \"Обробка природної мови – Ukrainian\")\n- [Tiếng Việt](https:\/\/vi.wikipedia.org\/wiki\/X%E1%BB%AD_l%C3%BD_ng%C3%B4n_ng%E1%BB%AF_t%E1%BB%B1_nhi%C3%AAn \"Xử lý ngôn ngữ tự nhiên – Vietnamese\")\n- [閩南語 \/ Bân-lâm-gí](https:\/\/zh-min-nan.wikipedia.org\/wiki\/Ch%C5%AB-ji%C3%A2n_gi%C3%A2n-g%C3%BA_chh%C3%BA-l%C3%AD \"Chū-jiân giân-gú chhú-lí – Minnan\")\n- [粵語](https:\/\/zh-yue.wikipedia.org\/wiki\/%E8%87%AA%E7%84%B6%E8%AA%9E%E8%A8%80%E8%99%95%E7%90%86 \"自然語言處理 – Cantonese\")\n- [中文](https:\/\/zh.wikipedia.org\/wiki\/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86 \"自然语言处理 – Chinese\")\n- [IsiZulu](https:\/\/zu.wikipedia.org\/wiki\/Ukudludlunga_ulimi_lwemvelo \"Ukudludlunga ulimi lwemvelo – Zulu\")\n\n[Edit links](https:\/\/www.wikidata.org\/wiki\/Special:EntityPage\/Q30642#sitelinks-wikipedia \"Edit interlanguage links\")\n\n- [Article](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing \"View the content page [c]\")\n- [Talk](https:\/\/en.wikipedia.org\/wiki\/Talk:Natural_language_processing \"Discuss improvements to the content page [t]\")\n\nEnglish\n\n- [Read](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing)\n- [Edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit \"Edit this page [e]\")\n- [View history](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=history \"Past revisions of this page [h]\")\n\nTools\n\nTools\n\nmove to sidebar\n\nhide\n\nActions\n\n- [Read](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing)\n- [Edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit \"Edit this page [e]\")\n- [View history](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=history)\n\nGeneral\n\n- [What links here](https:\/\/en.wikipedia.org\/wiki\/Special:WhatLinksHere\/Natural_language_processing \"List of all English Wikipedia pages containing links to this page [j]\")\n- [Related changes](https:\/\/en.wikipedia.org\/wiki\/Special:RecentChangesLinked\/Natural_language_processing \"Recent changes in pages linked from this page [k]\")\n- [Upload file](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:File_Upload_Wizard \"Upload files [u]\")\n- [Permanent link](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&oldid=1343173705 \"Permanent link to this revision of this page\")\n- [Page information](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=info \"More information about this page\")\n- [Cite this page](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:CiteThisPage&page=Natural_language_processing&id=1343173705&wpFormIdentifier=titleform \"Information on how to cite this page\")\n- [Get shortened URL](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:UrlShortener&url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FNatural_language_processing)\n\nPrint\/export\n\n- [Download as PDF](https:\/\/en.wikipedia.org\/w\/index.php?title=Special:DownloadAsPdf&page=Natural_language_processing&action=show-download-screen \"Download this page as a PDF file\")\n- [Printable version](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&printable=yes \"Printable version of this page [p]\")\n\nIn other projects\n\n- [Wikimedia Commons](https:\/\/commons.wikimedia.org\/wiki\/Category:Natural_language_processing)\n- [Wikiversity](https:\/\/en.wikiversity.org\/wiki\/Natural_Language_Processing)\n- [Wikidata item](https:\/\/www.wikidata.org\/wiki\/Special:EntityPage\/Q30642 \"Structured data on this page hosted by Wikidata [g]\")\n\nAppearance\n\nmove to sidebar\n\nhide\n\nFrom Wikipedia, the free encyclopedia\n\nProcessing of natural language by a computer\n\n|   |   |\n|---|---|\n| ![](https:\/\/upload.wikimedia.org\/wikipedia\/en\/thumb\/b\/b4\/Ambox_important.svg\/40px-Ambox_important.svg.png) |   |\n|   |   |\n| [![icon](https:\/\/upload.wikimedia.org\/wikipedia\/en\/thumb\/9\/99\/Question_book-new.svg\/60px-Question_book-new.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:Question_book-new.svg) | This article **needs additional citations for [verification](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Verifiability \"Wikipedia:Verifiability\")**. Please help [improve this article](https:\/\/en.wikipedia.org\/wiki\/Special:EditPage\/Natural_language_processing \"Special:EditPage\/Natural language processing\") by [adding citations to reliable sources](https:\/\/en.wikipedia.org\/wiki\/Help:Referencing_for_beginners \"Help:Referencing for beginners\"). Unsourced material may be challenged and removed. *Find sources:* [\"Natural language processing\"](https:\/\/www.google.com\/search?as_eq=wikipedia&q=%22Natural+language+processing%22) – [news](https:\/\/www.google.com\/search?tbm=nws&q=%22Natural+language+processing%22+-wikipedia&tbs=ar:1) **·** [newspapers](https:\/\/www.google.com\/search?&q=%22Natural+language+processing%22&tbs=bkt:s&tbm=bks) **·** [books](https:\/\/www.google.com\/search?tbs=bks:1&q=%22Natural+language+processing%22+-wikipedia) **·** [scholar](https:\/\/scholar.google.com\/scholar?q=%22Natural+language+processing%22) **·** [JSTOR](https:\/\/www.jstor.org\/action\/doBasicSearch?Query=%22Natural+language+processing%22&acc=on&wc=on) *(May 2024)* *([Learn how and when to remove this message](https:\/\/en.wikipedia.org\/wiki\/Help:Maintenance_template_removal \"Help:Maintenance template removal\"))* |\n|   |   |\n| [![icon](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/e\/e8\/Crystal_Clear_app_kedit.svg\/40px-Crystal_Clear_app_kedit.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:Crystal_Clear_app_kedit.svg) | This article **may need to be rewritten** to comply with Wikipedia's [quality standards](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Manual_of_Style \"Wikipedia:Manual of Style\"). Relevant discussion may be found on the [talk page](https:\/\/en.wikipedia.org\/wiki\/Talk:Natural_language_processing \"Talk:Natural language processing\"). [You can help](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit). The talk page may contain suggestions. *(July 2025)* |\n|   |   |\n| ![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/1\/1c\/Ambox_rewrite.svg\/40px-Ambox_rewrite.svg.png) | This article **may be in need of reorganization to comply with Wikipedia's [layout guidelines](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Manual_of_Style\/Layout \"Wikipedia:Manual of Style\/Layout\")**. Please help by [editing the article](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit) to make improvements to the overall structure. *(July 2025)* *([Learn how and when to remove this message](https:\/\/en.wikipedia.org\/wiki\/Help:Maintenance_template_removal \"Help:Maintenance template removal\"))* |\n\n|   |\n|---|\n| Part of [a series](https:\/\/en.wikipedia.org\/wiki\/Category:Artificial_intelligence \"Category:Artificial intelligence\") on |\n| [Artificial intelligence (AI)](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence \"Artificial intelligence\") |\n| [![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/6\/64\/Dall-e_3_%28jan_%2724%29_artificial_intelligence_icon.png\/120px-Dall-e_3_%28jan_%2724%29_artificial_intelligence_icon.png)](https:\/\/en.wikipedia.org\/wiki\/File:Dall-e_3_\\(jan_%2724\\)_artificial_intelligence_icon.png) |\n| [Major goals](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence#Goals \"Artificial intelligence\") [Artificial general intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_general_intelligence \"Artificial general intelligence\") [Intelligent agent](https:\/\/en.wikipedia.org\/wiki\/Intelligent_agent \"Intelligent agent\") [Recursive self-improvement](https:\/\/en.wikipedia.org\/wiki\/Recursive_self-improvement \"Recursive self-improvement\") [Planning](https:\/\/en.wikipedia.org\/wiki\/Automated_planning_and_scheduling \"Automated planning and scheduling\") [Computer vision](https:\/\/en.wikipedia.org\/wiki\/Computer_vision \"Computer vision\") [General game playing](https:\/\/en.wikipedia.org\/wiki\/General_game_playing \"General game playing\") [Knowledge representation](https:\/\/en.wikipedia.org\/wiki\/Knowledge_representation_and_reasoning \"Knowledge representation and reasoning\") [Natural language processing]() [Robotics](https:\/\/en.wikipedia.org\/wiki\/Robotics \"Robotics\") [AI safety](https:\/\/en.wikipedia.org\/wiki\/AI_safety \"AI safety\") |\n| Approaches [Machine learning](https:\/\/en.wikipedia.org\/wiki\/Machine_learning \"Machine learning\") [Symbolic](https:\/\/en.wikipedia.org\/wiki\/Symbolic_artificial_intelligence \"Symbolic artificial intelligence\") [Deep learning](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\") [Bayesian networks](https:\/\/en.wikipedia.org\/wiki\/Bayesian_network \"Bayesian network\") [Evolutionary algorithms](https:\/\/en.wikipedia.org\/wiki\/Evolutionary_algorithm \"Evolutionary algorithm\") [Hybrid intelligent systems](https:\/\/en.wikipedia.org\/wiki\/Hybrid_intelligent_system \"Hybrid intelligent system\") [Systems integration](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_systems_integration \"Artificial intelligence systems integration\") [Open-source](https:\/\/en.wikipedia.org\/wiki\/Open-source_artificial_intelligence \"Open-source artificial intelligence\") [AI data centers](https:\/\/en.wikipedia.org\/wiki\/AI_data_center \"AI data center\") |\n| [Applications](https:\/\/en.wikipedia.org\/wiki\/Applications_of_artificial_intelligence \"Applications of artificial intelligence\") [Bioinformatics](https:\/\/en.wikipedia.org\/wiki\/Machine_learning_in_bioinformatics \"Machine learning in bioinformatics\") [Deepfake](https:\/\/en.wikipedia.org\/wiki\/Deepfake \"Deepfake\") [Earth sciences](https:\/\/en.wikipedia.org\/wiki\/Machine_learning_in_earth_sciences \"Machine learning in earth sciences\") [Finance](https:\/\/en.wikipedia.org\/wiki\/Applications_of_artificial_intelligence#Finance \"Applications of artificial intelligence\") [Generative AI](https:\/\/en.wikipedia.org\/wiki\/Generative_artificial_intelligence \"Generative artificial intelligence\") [Art](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_art \"Artificial intelligence art\") [Audio](https:\/\/en.wikipedia.org\/wiki\/Generative_audio \"Generative audio\") [Music](https:\/\/en.wikipedia.org\/wiki\/Music_and_artificial_intelligence \"Music and artificial intelligence\") [Government](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_government \"Artificial intelligence in government\") [Healthcare](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_healthcare \"Artificial intelligence in healthcare\") [Mental health](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_mental_health \"Artificial intelligence in mental health\") [Industry](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_industry \"Artificial intelligence in industry\") [Software development](https:\/\/en.wikipedia.org\/wiki\/AI-assisted_software_development \"AI-assisted software development\") [Translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\") [Military](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_arms_race \"Artificial intelligence arms race\") [Physics](https:\/\/en.wikipedia.org\/wiki\/Machine_learning_in_physics \"Machine learning in physics\") [Projects](https:\/\/en.wikipedia.org\/wiki\/List_of_artificial_intelligence_projects \"List of artificial intelligence projects\") |\n| [Philosophy](https:\/\/en.wikipedia.org\/wiki\/Philosophy_of_artificial_intelligence \"Philosophy of artificial intelligence\") [AI alignment](https:\/\/en.wikipedia.org\/wiki\/AI_alignment \"AI alignment\") [Artificial consciousness](https:\/\/en.wikipedia.org\/wiki\/Artificial_consciousness \"Artificial consciousness\") [The bitter lesson](https:\/\/en.wikipedia.org\/wiki\/Bitter_lesson \"Bitter lesson\") [Chinese room](https:\/\/en.wikipedia.org\/wiki\/Chinese_room \"Chinese room\") [Friendly AI](https:\/\/en.wikipedia.org\/wiki\/Friendly_artificial_intelligence \"Friendly artificial intelligence\") [Ethics](https:\/\/en.wikipedia.org\/wiki\/Ethics_of_artificial_intelligence \"Ethics of artificial intelligence\") [Existential risk](https:\/\/en.wikipedia.org\/wiki\/Existential_risk_from_artificial_general_intelligence \"Existential risk from artificial general intelligence\") [Turing test](https:\/\/en.wikipedia.org\/wiki\/Turing_test \"Turing test\") [Uncanny valley](https:\/\/en.wikipedia.org\/wiki\/Uncanny_valley \"Uncanny valley\") [Human–AI interaction](https:\/\/en.wikipedia.org\/wiki\/Human%E2%80%93AI_interaction \"Human–AI interaction\") |\n| [History](https:\/\/en.wikipedia.org\/wiki\/History_of_artificial_intelligence \"History of artificial intelligence\") [Timeline](https:\/\/en.wikipedia.org\/wiki\/Timeline_of_artificial_intelligence \"Timeline of artificial intelligence\") [Progress](https:\/\/en.wikipedia.org\/wiki\/Progress_in_artificial_intelligence \"Progress in artificial intelligence\") [AI winter](https:\/\/en.wikipedia.org\/wiki\/AI_winter \"AI winter\") [AI boom](https:\/\/en.wikipedia.org\/wiki\/AI_boom \"AI boom\") [AI bubble](https:\/\/en.wikipedia.org\/wiki\/AI_bubble \"AI bubble\") |\n| [Controversies](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_controversies \"Artificial intelligence controversies\") [Deepfake pornography](https:\/\/en.wikipedia.org\/wiki\/Deepfake_pornography \"Deepfake pornography\") [Taylor Swift deepfake pornography controversy](https:\/\/en.wikipedia.org\/wiki\/Taylor_Swift_deepfake_pornography_controversy \"Taylor Swift deepfake pornography controversy\") [Grok sexual deepfake scandal](https:\/\/en.wikipedia.org\/wiki\/Grok_sexual_deepfake_scandal \"Grok sexual deepfake scandal\") [Google Gemini image generation controversy](https:\/\/en.wikipedia.org\/wiki\/Google_Gemini_image_generation_controversy \"Google Gemini image generation controversy\") *[It's the Most Terrible Time of the Year](https:\/\/en.wikipedia.org\/wiki\/It%27s_the_Most_Terrible_Time_of_the_Year \"It's the Most Terrible Time of the Year\")* [Pause Giant AI Experiments](https:\/\/en.wikipedia.org\/wiki\/Pause_Giant_AI_Experiments:_An_Open_Letter \"Pause Giant AI Experiments: An Open Letter\") [Removal of Sam Altman from OpenAI](https:\/\/en.wikipedia.org\/wiki\/Removal_of_Sam_Altman_from_OpenAI \"Removal of Sam Altman from OpenAI\") [Statement on AI Risk](https:\/\/en.wikipedia.org\/wiki\/Statement_on_AI_Risk \"Statement on AI Risk\") [Tay (chatbot)](https:\/\/en.wikipedia.org\/wiki\/Tay_\\(chatbot\\) \"Tay (chatbot)\") *[Théâtre D'opéra Spatial](https:\/\/en.wikipedia.org\/wiki\/Th%C3%A9%C3%A2tre_D%27op%C3%A9ra_Spatial \"Théâtre D'opéra Spatial\")* [Voiceverse NFT plagiarism scandal](https:\/\/en.wikipedia.org\/wiki\/Voiceverse_NFT_plagiarism_scandal \"Voiceverse NFT plagiarism scandal\") |\n| Glossary [Glossary](https:\/\/en.wikipedia.org\/wiki\/Glossary_of_artificial_intelligence \"Glossary of artificial intelligence\") |\n| [v](https:\/\/en.wikipedia.org\/wiki\/Template:Artificial_intelligence \"Template:Artificial intelligence\") [t](https:\/\/en.wikipedia.org\/wiki\/Template_talk:Artificial_intelligence \"Template talk:Artificial intelligence\") [e](https:\/\/en.wikipedia.org\/wiki\/Special:EditPage\/Template:Artificial_intelligence \"Special:EditPage\/Template:Artificial intelligence\") |\n\n**Natural language processing** (**NLP**) is the processing of [natural language](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") information by a [computer](https:\/\/en.wikipedia.org\/wiki\/Computer \"Computer\"). NLP is a subfield of [computer science](https:\/\/en.wikipedia.org\/wiki\/Computer_science \"Computer science\") and is closely associated with [artificial intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence \"Artificial intelligence\"). NLP is also related to [information retrieval](https:\/\/en.wikipedia.org\/wiki\/Information_retrieval \"Information retrieval\"), [knowledge representation](https:\/\/en.wikipedia.org\/wiki\/Knowledge_representation_and_reasoning \"Knowledge representation and reasoning\"), [computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\"), and [linguistics](https:\/\/en.wikipedia.org\/wiki\/Linguistics \"Linguistics\") more broadly.[\\[1\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-nlpintro-1)\n\nMajor processing tasks in an NLP system include: [speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\"), [text classification](https:\/\/en.wikipedia.org\/wiki\/Text_classification \"Text classification\"), [natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_understanding \"Natural language understanding\"), and [natural language generation](https:\/\/en.wikipedia.org\/wiki\/Natural_language_generation \"Natural language generation\").\n\n## History\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=1 \"Edit section: History\")\\]\n\nFurther information: [History of natural language processing](https:\/\/en.wikipedia.org\/wiki\/History_of_natural_language_processing \"History of natural language processing\")\n\nNatural language processing has its roots in the 1950s.[\\[2\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-2) Already in 1950, [Alan Turing](https:\/\/en.wikipedia.org\/wiki\/Alan_Turing \"Alan Turing\") published an article titled \"[Computing Machinery and Intelligence](https:\/\/en.wikipedia.org\/wiki\/Computing_Machinery_and_Intelligence \"Computing Machinery and Intelligence\")\" which proposed what is now called the [Turing test](https:\/\/en.wikipedia.org\/wiki\/Turing_test \"Turing test\") as a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language.\n\n### Symbolic NLP (1950s – early 1990s)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=2 \"Edit section: Symbolic NLP (1950s – early 1990s)\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/0\/09\/AST_Document.svg\/250px-AST_Document.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:AST_Document.svg)\n\nA document parsed into an abstract syntax tree\n\nThe premise of symbolic NLP is often illustrated using [John Searle's Chinese room](https:\/\/en.wikipedia.org\/wiki\/John_Searle_\\(American_philosopher\\) \"John Searle (American philosopher)\") thought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts.\n\n- **1950s**: The [Georgetown experiment](https:\/\/en.wikipedia.org\/wiki\/Georgetown-IBM_experiment \"Georgetown-IBM experiment\") in 1954 involved fully [automatic translation](https:\/\/en.wikipedia.org\/wiki\/Automatic_translation \"Automatic translation\") of more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem.[\\[3\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-3) However, real progress was much slower, and after the [ALPAC report](https:\/\/en.wikipedia.org\/wiki\/ALPAC \"ALPAC\") in 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe[\\[4\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-4)) until the late 1980s when the first [statistical machine translation](https:\/\/en.wikipedia.org\/wiki\/Statistical_machine_translation \"Statistical machine translation\") systems were developed.\n- **1960s**: Some notably successful natural language processing systems developed in the 1960s were [SHRDLU](https:\/\/en.wikipedia.org\/wiki\/SHRDLU \"SHRDLU\"), a natural language system working in restricted \"[blocks worlds](https:\/\/en.wikipedia.org\/wiki\/Blocks_world \"Blocks world\")\" with restricted vocabularies, and [ELIZA](https:\/\/en.wikipedia.org\/wiki\/ELIZA \"ELIZA\"), a simulation of a [Rogerian psychotherapy](https:\/\/en.wikipedia.org\/wiki\/Rogerian_psychotherapy \"Rogerian psychotherapy\"), written by [Joseph Weizenbaum](https:\/\/en.wikipedia.org\/wiki\/Joseph_Weizenbaum \"Joseph Weizenbaum\") between 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the \"patient\" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to \"My head hurts\" with \"Why do you say your head hurts?\". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only *twenty* words, because that was all that would fit in a computer memory at the time.[\\[5\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-5)\n- **1970s**: During the 1970s, many programmers began to write \"conceptual [ontologies](https:\/\/en.wikipedia.org\/wiki\/Ontology_\\(information_science\\) \"Ontology (information science)\")\", which structured real-world information into computer-understandable data. Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981). During this time, the first [chatterbots](https:\/\/en.wikipedia.org\/wiki\/Chatterbots \"Chatterbots\") were written (e.g., [PARRY](https:\/\/en.wikipedia.org\/wiki\/PARRY \"PARRY\")).\n- **1980s**: The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of [HPSG](https:\/\/en.wikipedia.org\/wiki\/Head-driven_phrase_structure_grammar \"Head-driven phrase structure grammar\") as a computational operationalization of [generative grammar](https:\/\/en.wikipedia.org\/wiki\/Generative_grammar \"Generative grammar\")), morphology (e.g., two-level morphology[\\[6\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-6)), semantics (e.g., [Lesk algorithm](https:\/\/en.wikipedia.org\/wiki\/Lesk_algorithm \"Lesk algorithm\")), reference (e.g., within Centering Theory[\\[7\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-7)) and other areas of natural language understanding (e.g., in the [Rhetorical Structure Theory](https:\/\/en.wikipedia.org\/wiki\/Rhetorical_structure_theory \"Rhetorical structure theory\")). Other lines of research were continued, e.g., the development of chatterbots with [Racter](https:\/\/en.wikipedia.org\/wiki\/Racter \"Racter\") and [Jabberwacky](https:\/\/en.wikipedia.org\/wiki\/Jabberwacky \"Jabberwacky\"). An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period.[\\[8\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-8)\n\n### Statistical NLP (1990s–present)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=3 \"Edit section: Statistical NLP (1990s–present)\")\\]\n\nUp until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of [machine learning](https:\/\/en.wikipedia.org\/wiki\/Machine_learning \"Machine learning\") algorithms for language processing. This shift was influenced by increasing computational power (see [Moore's law](https:\/\/en.wikipedia.org\/wiki\/Moore%27s_law \"Moore's law\")) and a decline in the dominance of [Chomskyan](https:\/\/en.wikipedia.org\/wiki\/Chomskyan \"Chomskyan\") linguistic theories... (e.g. [transformational grammar](https:\/\/en.wikipedia.org\/wiki\/Transformational_grammar \"Transformational grammar\")), whose theoretical underpinnings discouraged the sort of [corpus linguistics](https:\/\/en.wikipedia.org\/wiki\/Corpus_linguistics \"Corpus linguistics\") that underlies the machine-learning approach to language processing.[\\[9\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-9)\n\n- **1990s**: Many of the notable early successes in statistical methods in NLP occurred in the field of [machine translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\"), due especially to work at IBM Research, such as [IBM alignment models](https:\/\/en.wikipedia.org\/wiki\/IBM_alignment_models \"IBM alignment models\"). These systems were able to take advantage of existing multilingual [textual corpora](https:\/\/en.wikipedia.org\/wiki\/Text_corpus \"Text corpus\") that had been produced by the [Parliament of Canada](https:\/\/en.wikipedia.org\/wiki\/Parliament_of_Canada \"Parliament of Canada\") and the [European Union](https:\/\/en.wikipedia.org\/wiki\/European_Union \"European Union\") as a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data.\n- **2000s**: With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on [unsupervised](https:\/\/en.wikipedia.org\/wiki\/Unsupervised_learning \"Unsupervised learning\") and [semi-supervised learning](https:\/\/en.wikipedia.org\/wiki\/Semi-supervised_learning \"Semi-supervised learning\") algorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than [supervised learning](https:\/\/en.wikipedia.org\/wiki\/Supervised_learning \"Supervised learning\"), and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the [World Wide Web](https:\/\/en.wikipedia.org\/wiki\/World_Wide_Web \"World Wide Web\")), which can often make up for the worse efficiency if the algorithm used has a low enough [time complexity](https:\/\/en.wikipedia.org\/wiki\/Time_complexity \"Time complexity\") to be practical.\n- **2003:** [word n-gram model](https:\/\/en.wikipedia.org\/wiki\/Word_n-gram_language_model \"Word n-gram language model\"), at the time the best statistical algorithm, is outperformed by a [multi-layer perceptron](https:\/\/en.wikipedia.org\/wiki\/Multi-layer_perceptron \"Multi-layer perceptron\") (with a single hidden layer and context length of several words, trained on up to 14 million words, by [Bengio](https:\/\/en.wikipedia.org\/wiki\/Yoshua_Bengio \"Yoshua Bengio\") et al.)[\\[10\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-10)\n- **2010:** [Tomáš Mikolov](https:\/\/en.wikipedia.org\/wiki\/Tom%C3%A1%C5%A1_Mikolov \"Tomáš Mikolov\") (then a PhD student at [Brno University of Technology](https:\/\/en.wikipedia.org\/wiki\/Brno_University_of_Technology \"Brno University of Technology\")) with co-authors applied a simple [recurrent neural network](https:\/\/en.wikipedia.org\/wiki\/Recurrent_neural_network \"Recurrent neural network\") with a single hidden layer to language modeling,[\\[11\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-11) and in the following years he went on to develop [Word2vec](https:\/\/en.wikipedia.org\/wiki\/Word2vec \"Word2vec\"). In the 2010s, [representation learning](https:\/\/en.wikipedia.org\/wiki\/Representation_learning \"Representation learning\") and [deep neural network](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\")\\-style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques[\\[12\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-goldberg:nnlp17-12)[\\[13\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-goodfellow:book16-13) can achieve state-of-the-art results in many natural language tasks, e.g., in [language modeling](https:\/\/en.wikipedia.org\/wiki\/Language_modeling \"Language modeling\")[\\[14\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-jozefowicz:lm16-14) and parsing.[\\[15\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-choe:emnlp16-15)[\\[16\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-vinyals:nips15-16) This is increasingly important [in medicine and healthcare](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_healthcare \"Artificial intelligence in healthcare\"), where NLP helps analyze notes and text in [electronic health records](https:\/\/en.wikipedia.org\/wiki\/Electronic_health_record \"Electronic health record\") that would otherwise be inaccessible for study when seeking to improve care[\\[17\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-17) or protect patient privacy.[\\[18\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-18)\n\n## Approaches: Symbolic, statistical, neural networks\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=4 \"Edit section: Approaches: Symbolic, statistical, neural networks\")\\]\n\nSymbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular:[\\[19\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-winograd:shrdlu71-19)[\\[20\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-schank77-20) such as by writing grammars or devising heuristic rules for [stemming](https:\/\/en.wikipedia.org\/wiki\/Stemming \"Stemming\").\n\n[Machine learning](https:\/\/en.wikipedia.org\/wiki\/Machine_learning \"Machine learning\") approaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach:\n\n- both statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally.\n- [language models](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\"), produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce.\n- the larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to [intractability](https:\/\/en.wikipedia.org\/wiki\/Intractable_problem \"Intractable problem\") problems.\n\nRule-based systems are commonly used:\n\n- when the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the [Apertium](https:\/\/en.wikipedia.org\/wiki\/Apertium \"Apertium\") system,\n- for preprocessing in NLP pipelines, e.g., [tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(lexical_analysis\\) \"Tokenization (lexical analysis)\"), or\n- for post-processing and transforming the output of NLP pipelines, e.g., for [knowledge extraction](https:\/\/en.wikipedia.org\/wiki\/Knowledge_extraction \"Knowledge extraction\") from syntactic parses.\n\n### Statistical approach\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=5 \"Edit section: Statistical approach\")\\]\n\nIn the late 1980s and mid-1990s, the statistical approach ended a period of [AI winter](https:\/\/en.wikipedia.org\/wiki\/AI_winter \"AI winter\"), which was caused by the inefficiencies of the rule-based approaches.[\\[21\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-johnson:eacl:ilcl09-21)[\\[22\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-resnik:langlog11-22)\n\nThe earliest [decision trees](https:\/\/en.wikipedia.org\/wiki\/Decision_tree \"Decision tree\"), producing systems of hard [if–then rules](https:\/\/en.wikipedia.org\/wiki\/Conditional_\\(computer_programming\\)#If%E2%80%93then\\(%E2%80%93else\\) \"Conditional (computer programming)\"), were still very similar to the old rule-based approaches. Only the introduction of hidden [Markov models](https:\/\/en.wikipedia.org\/wiki\/Markov_model \"Markov model\"), applied to part-of-speech tagging, announced the end of the old rule-based approach.\n\n### Neural networks\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=6 \"Edit section: Neural networks\")\\]\n\nFurther information: [Artificial neural network](https:\/\/en.wikipedia.org\/wiki\/Artificial_neural_network \"Artificial neural network\")\n\nA major drawback of statistical methods is that they require elaborate [feature engineering](https:\/\/en.wikipedia.org\/wiki\/Feature_engineering \"Feature engineering\"). Since 2015,[\\[23\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-23) [neural network](https:\/\/en.wikipedia.org\/wiki\/Neural_network_\\(machine_learning\\) \"Neural network (machine learning)\")–based methods have increasingly replaced traditional statistical approaches, using [semantic networks](https:\/\/en.wikipedia.org\/wiki\/Semantic_networks \"Semantic networks\")[\\[24\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-24) and [word embeddings](https:\/\/en.wikipedia.org\/wiki\/Word_embedding \"Word embedding\") to capture semantic properties of words.\n\nIntermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore.\n\n[Neural machine translation](https:\/\/en.wikipedia.org\/wiki\/Neural_machine_translation \"Neural machine translation\"), based on then-newly invented [sequence-to-sequence](https:\/\/en.wikipedia.org\/wiki\/Seq2seq \"Seq2seq\") transformations, made obsolete the intermediate steps, such as word alignment, previously necessary for [statistical machine translation](https:\/\/en.wikipedia.org\/wiki\/Statistical_machine_translation \"Statistical machine translation\").\n\n## Common NLP tasks\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=7 \"Edit section: Common NLP tasks\")\\]\n\nThe following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks.\n\nThough natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below.\n\n### Text and speech processing\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=8 \"Edit section: Text and speech processing\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/8\/8c\/Hebrew_stop_words_word_cloud.png\/250px-Hebrew_stop_words_word_cloud.png)](https:\/\/en.wikipedia.org\/wiki\/File:Hebrew_stop_words_word_cloud.png)\n\nWord cloud of stop words in Hebrew\n\n[Optical character recognition](https:\/\/en.wikipedia.org\/wiki\/Optical_character_recognition \"Optical character recognition\") (OCR)\n\nGiven an image representing printed text, determine the corresponding text.\n\n[Speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\")\n\nGiven a sound clip of a person or people speaking, determine the textual representation of the speech. This is the opposite of [text to speech](https:\/\/en.wikipedia.org\/wiki\/Text_to_speech \"Text to speech\") and is one of the extremely difficult problems colloquially termed \"[AI-complete](https:\/\/en.wikipedia.org\/wiki\/AI-complete \"AI-complete\")\" (see above). In [natural speech](https:\/\/en.wikipedia.org\/wiki\/Natural_speech \"Natural speech\") there are hardly any pauses between successive words, and thus [speech segmentation](https:\/\/en.wikipedia.org\/wiki\/Speech_segmentation \"Speech segmentation\") is a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed [coarticulation](https:\/\/en.wikipedia.org\/wiki\/Coarticulation \"Coarticulation\"), so the conversion of the [analog signal](https:\/\/en.wikipedia.org\/wiki\/Analog_signal \"Analog signal\") to discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent.\n\n[Speech segmentation](https:\/\/en.wikipedia.org\/wiki\/Speech_segmentation \"Speech segmentation\")\n\nGiven a sound clip of a person or people speaking, separate it into words. A subtask of [speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\") and typically grouped with it.\n\n[Text-to-speech](https:\/\/en.wikipedia.org\/wiki\/Text-to-speech \"Text-to-speech\")\n\nGiven a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired.[\\[25\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-25)\n\n[Word segmentation](https:\/\/en.wikipedia.org\/wiki\/Word_segmentation \"Word segmentation\") ([Tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(lexical_analysis\\) \"Tokenization (lexical analysis)\"))\n\n[Tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(data_security\\) \"Tokenization (data security)\") is a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods.[\\[26\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-:0-26)\n\nFor a language like [English](https:\/\/en.wikipedia.org\/wiki\/English_language \"English language\"), this is fairly trivial, since words are usually separated by spaces. However, some written languages like [Chinese](https:\/\/en.wikipedia.org\/wiki\/Chinese_language \"Chinese language\"), [Japanese](https:\/\/en.wikipedia.org\/wiki\/Japanese_language \"Japanese language\") and [Thai](https:\/\/en.wikipedia.org\/wiki\/Thai_language \"Thai language\") do not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the [vocabulary](https:\/\/en.wikipedia.org\/wiki\/Vocabulary \"Vocabulary\") and [morphology](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\") of words in the language. Sometimes this process is also used in cases like [bag of words](https:\/\/en.wikipedia.org\/wiki\/Bag_of_words \"Bag of words\") (BOW) creation in data mining.[\\[27\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-27)[\\[28\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-28)\n\n### Morphological analysis\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=9 \"Edit section: Morphological analysis\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/2\/27\/Eustagger_euskal_lematizatzailearen_adibidea.png\/250px-Eustagger_euskal_lematizatzailearen_adibidea.png)](https:\/\/en.wikipedia.org\/wiki\/File:Eustagger_euskal_lematizatzailearen_adibidea.png)\n\nLemmatization of Basque words\n\n[Lemmatization](https:\/\/en.wikipedia.org\/wiki\/Lemmatisation \"Lemmatisation\")\n\nThe task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form.[\\[29\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-29)\n\n[Morphological segmentation](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\")\n\nSeparate words into individual [morphemes](https:\/\/en.wikipedia.org\/wiki\/Morpheme \"Morpheme\") and identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the [morphology](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\") (*i.e.*, the structure of words) of the language being considered. [English](https:\/\/en.wikipedia.org\/wiki\/English_language \"English language\") has fairly simple morphology, especially [inflectional morphology](https:\/\/en.wikipedia.org\/wiki\/Inflectional_morphology \"Inflectional morphology\"), and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., \"open, opens, opened, opening\") as separate words. In languages such as [Turkish](https:\/\/en.wikipedia.org\/wiki\/Turkish_language \"Turkish language\") or [Meitei](https:\/\/en.wikipedia.org\/wiki\/Meitei_language \"Meitei language\"), a highly [agglutinated](https:\/\/en.wikipedia.org\/wiki\/Agglutination \"Agglutination\") Indian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms.[\\[30\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-30)\n\n[Part-of-speech tagging](https:\/\/en.wikipedia.org\/wiki\/Part-of-speech_tagging \"Part-of-speech tagging\")\n\nGiven a sentence, determine the [part of speech](https:\/\/en.wikipedia.org\/wiki\/Part_of_speech \"Part of speech\") (POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, \"book\" can be a [noun](https:\/\/en.wikipedia.org\/wiki\/Noun \"Noun\") (\"the book on the table\") or [verb](https:\/\/en.wikipedia.org\/wiki\/Verb \"Verb\") (\"to book a flight\"); \"set\" can be a noun, verb or [adjective](https:\/\/en.wikipedia.org\/wiki\/Adjective \"Adjective\"); and \"out\" can be any of at least five different parts of speech.\n\n[Stemming](https:\/\/en.wikipedia.org\/wiki\/Stemming \"Stemming\")\n\nThe process of reducing inflected (or sometimes derived) words to a base form (e.g., \"close\" will be the root for \"closed\", \"closing\", \"close\", \"closer\" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary.\n\n### Syntactic analysis\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=10 \"Edit section: Syntactic analysis\")\\]\n\n|   |\n|---|\n| Part of [*a series*](https:\/\/en.wikipedia.org\/wiki\/Category:Formal_languages \"Category:Formal languages\") on |\n| [Formal languages](https:\/\/en.wikipedia.org\/wiki\/Formal_languages \"Formal languages\") |\n| Key concepts [Formal system](https:\/\/en.wikipedia.org\/wiki\/Formal_system \"Formal system\") [Alphabet](https:\/\/en.wikipedia.org\/wiki\/Alphabet_\\(formal_languages\\) \"Alphabet (formal languages)\") [Syntax](https:\/\/en.wikipedia.org\/wiki\/Syntax_\\(logic\\) \"Syntax (logic)\") [Formal semantics](https:\/\/en.wikipedia.org\/wiki\/Semantics_of_logic \"Semantics of logic\") [Semantics (programming languages)](https:\/\/en.wikipedia.org\/wiki\/Semantics_\\(computer_science\\) \"Semantics (computer science)\") [Formal grammar](https:\/\/en.wikipedia.org\/wiki\/Formal_grammar \"Formal grammar\") [Formation rule](https:\/\/en.wikipedia.org\/wiki\/Formation_rule \"Formation rule\") [Well-formed formula](https:\/\/en.wikipedia.org\/wiki\/Well-formed_formula \"Well-formed formula\") [Automata theory](https:\/\/en.wikipedia.org\/wiki\/Automata_theory \"Automata theory\") [Regular expression](https:\/\/en.wikipedia.org\/wiki\/Regular_expression \"Regular expression\") [Production](https:\/\/en.wikipedia.org\/wiki\/Production_\\(computer_science\\) \"Production (computer science)\") [Ground expression](https:\/\/en.wikipedia.org\/wiki\/Ground_expression \"Ground expression\") [Atomic formula](https:\/\/en.wikipedia.org\/wiki\/Atomic_formula \"Atomic formula\") |\n| Applications [Formal methods](https:\/\/en.wikipedia.org\/wiki\/Formal_methods \"Formal methods\") [Propositional calculus](https:\/\/en.wikipedia.org\/wiki\/Propositional_calculus \"Propositional calculus\") [Predicate logic](https:\/\/en.wikipedia.org\/wiki\/Predicate_logic \"Predicate logic\") [Mathematical notation](https:\/\/en.wikipedia.org\/wiki\/Mathematical_notation \"Mathematical notation\") [Natural language processing]() [Programming language theory](https:\/\/en.wikipedia.org\/wiki\/Programming_language_theory \"Programming language theory\") [Mathematical linguistics](https:\/\/en.wikipedia.org\/wiki\/Mathematical_linguistics \"Mathematical linguistics\") [Computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\") [Syntax analysis](https:\/\/en.wikipedia.org\/wiki\/Syntax_analysis \"Syntax analysis\") [Formal verification](https:\/\/en.wikipedia.org\/wiki\/Formal_verification \"Formal verification\") [Automated theorem proving](https:\/\/en.wikipedia.org\/wiki\/Automated_theorem_proving \"Automated theorem proving\") |\n| [v](https:\/\/en.wikipedia.org\/wiki\/Template:Formal_languages \"Template:Formal languages\") [t](https:\/\/en.wikipedia.org\/w\/index.php?title=Template_talk:Formal_languages&action=edit&redlink=1 \"Template talk:Formal languages (page does not exist)\") [e](https:\/\/en.wikipedia.org\/wiki\/Special:EditPage\/Template:Formal_languages \"Special:EditPage\/Template:Formal languages\") |\n\n[Grammar induction](https:\/\/en.wikipedia.org\/wiki\/Grammar_induction \"Grammar induction\")[\\[31\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-31)\n\nGenerate a [formal grammar](https:\/\/en.wikipedia.org\/wiki\/Formal_grammar \"Formal grammar\") that describes a language's syntax.\n\n[Sentence breaking](https:\/\/en.wikipedia.org\/wiki\/Sentence_breaking \"Sentence breaking\") (also known as \"[sentence boundary disambiguation](https:\/\/en.wikipedia.org\/wiki\/Sentence_boundary_disambiguation \"Sentence boundary disambiguation\")\")\n\nGiven a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by [periods](https:\/\/en.wikipedia.org\/wiki\/Full_stop \"Full stop\") or other [punctuation marks](https:\/\/en.wikipedia.org\/wiki\/Punctuation_mark \"Punctuation mark\"), but these same characters can serve other purposes (e.g., marking [abbreviations](https:\/\/en.wikipedia.org\/wiki\/Abbreviation \"Abbreviation\")).\n\n[Parsing](https:\/\/en.wikipedia.org\/wiki\/Parsing \"Parsing\")\n\nDetermine the [parse tree](https:\/\/en.wikipedia.org\/wiki\/Parse_tree \"Parse tree\") (grammatical analysis) of a given sentence. The [grammar](https:\/\/en.wikipedia.org\/wiki\/Grammar \"Grammar\") for [natural languages](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") is [ambiguous](https:\/\/en.wikipedia.org\/wiki\/Ambiguous \"Ambiguous\") and typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing: *dependency parsing* and *constituency parsing*. Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a [probabilistic context-free grammar](https:\/\/en.wikipedia.org\/wiki\/Probabilistic_context-free_grammar \"Probabilistic context-free grammar\") (PCFG) (see also *[stochastic grammar](https:\/\/en.wikipedia.org\/wiki\/Stochastic_grammar \"Stochastic grammar\")*).\n\n### Lexical semantics (of individual words in context)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=11 \"Edit section: Lexical semantics (of individual words in context)\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/3\/34\/Entity_Linking_-_Example_of_pipeline.png\/250px-Entity_Linking_-_Example_of_pipeline.png)](https:\/\/en.wikipedia.org\/wiki\/File:Entity_Linking_-_Example_of_pipeline.png)\n\nAn entity linking pipeline\n\n[Lexical semantics](https:\/\/en.wikipedia.org\/wiki\/Lexical_semantics \"Lexical semantics\")\n\nWhat is the computational meaning of individual words in context?\n\n[Distributional semantics](https:\/\/en.wikipedia.org\/wiki\/Distributional_semantics \"Distributional semantics\")\n\nHow can we learn semantic representations from data?\n\n[Named entity recognition](https:\/\/en.wikipedia.org\/wiki\/Named_entity_recognition \"Named entity recognition\") (NER)\n\nGiven a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although [capitalization](https:\/\/en.wikipedia.org\/wiki\/Capitalization \"Capitalization\") can aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of [named entity](https:\/\/en.wikipedia.org\/wiki\/Named_entity \"Named entity\"), and in any case, is often inaccurate or insufficient. For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized. Furthermore, many other languages in non-Western scripts (e.g. [Chinese](https:\/\/en.wikipedia.org\/wiki\/Chinese_language \"Chinese language\") or [Arabic](https:\/\/en.wikipedia.org\/wiki\/Arabic_language \"Arabic language\")) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example, [German](https:\/\/en.wikipedia.org\/wiki\/German_language \"German language\") capitalizes all [nouns](https:\/\/en.wikipedia.org\/wiki\/Noun \"Noun\"), regardless of whether they are names, and [French](https:\/\/en.wikipedia.org\/wiki\/French_language \"French language\") and [Spanish](https:\/\/en.wikipedia.org\/wiki\/Spanish_language \"Spanish language\") do not capitalize names that serve as [adjectives](https:\/\/en.wikipedia.org\/wiki\/Adjective \"Adjective\"). This task is also referred to as token classification.[\\[32\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-32)\n\n[Sentiment analysis](https:\/\/en.wikipedia.org\/wiki\/Sentiment_analysis \"Sentiment analysis\") (see also [Multimodal sentiment analysis](https:\/\/en.wikipedia.org\/wiki\/Multimodal_sentiment_analysis \"Multimodal sentiment analysis\"))\n\nSentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as [word n-grams](https:\/\/en.wikipedia.org\/wiki\/Word_n-gram_language_model \"Word n-gram language model\"), [Term Frequency-Inverse Document Frequency](https:\/\/en.wikipedia.org\/wiki\/Term_frequency-inverse_document_frequency \"Term frequency-inverse document frequency\") (TF-IDF) features, hand-generated features, or employ [deep learning](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\") models designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms.[\\[26\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-:0-26)\n\n[Terminology extraction](https:\/\/en.wikipedia.org\/wiki\/Terminology_extraction \"Terminology extraction\")\n\nThe goal of terminology extraction is to automatically extract relevant terms from a given corpus.\n\n[Word-sense disambiguation](https:\/\/en.wikipedia.org\/wiki\/Word-sense_disambiguation \"Word-sense disambiguation\") (WSD)\n\nMany words have more than one [meaning](https:\/\/en.wikipedia.org\/wiki\/Meaning_\\(linguistics\\) \"Meaning (linguistics)\"); we have to select the meaning which makes the most sense in context. For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as [WordNet](https:\/\/en.wikipedia.org\/wiki\/WordNet \"WordNet\").\n\n[Entity linking](https:\/\/en.wikipedia.org\/wiki\/Entity_linking \"Entity linking\")\n\nMany words—typically proper names—refer to [named entities](https:\/\/en.wikipedia.org\/wiki\/Named_entity \"Named entity\"); here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context.\n\n### Relational semantics (semantics of individual sentences)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=12 \"Edit section: Relational semantics (semantics of individual sentences)\")\\]\n\n[Relationship extraction](https:\/\/en.wikipedia.org\/wiki\/Relationship_extraction \"Relationship extraction\")\n\nGiven a chunk of text, identify the relationships among named entities (e.g. who is married to whom).\n\n[Semantic parsing](https:\/\/en.wikipedia.org\/wiki\/Semantic_parsing \"Semantic parsing\")\n\nGiven a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in [AMR parsing](https:\/\/en.wikipedia.org\/wiki\/Abstract_Meaning_Representation \"Abstract Meaning Representation\")) or in accordance with a logical formalism (e.g., in [DRT parsing](https:\/\/en.wikipedia.org\/wiki\/Discourse_representation_theory \"Discourse representation theory\")). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see [Natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Natural_language_understanding) below).\n\n[Semantic role labelling](https:\/\/en.wikipedia.org\/wiki\/Semantic_role_labeling \"Semantic role labeling\") (see also implicit semantic role labelling below)\n\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https:\/\/en.wikipedia.org\/wiki\/Frame_semantics_\\(linguistics\\) \"Frame semantics (linguistics)\")), then identify and classify the frame elements ([semantic roles](https:\/\/en.wikipedia.org\/wiki\/Semantic_roles \"Semantic roles\")).\n\n### Discourse (semantics beyond individual sentences)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=13 \"Edit section: Discourse (semantics beyond individual sentences)\")\\]\n\n[Coreference resolution](https:\/\/en.wikipedia.org\/wiki\/Coreference \"Coreference\")\n\nGiven a sentence or larger chunk of text, determine which words (\"mentions\") refer to the same objects (\"entities\"). [Anaphora resolution](https:\/\/en.wikipedia.org\/wiki\/Anaphora_resolution \"Anaphora resolution\") is a specific example of this task, and is specifically concerned with matching up [pronouns](https:\/\/en.wikipedia.org\/wiki\/Pronoun \"Pronoun\") with the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called \"bridging relationships\" involving [referring expressions](https:\/\/en.wikipedia.org\/wiki\/Referring_expression \"Referring expression\"). For example, in a sentence such as \"He entered John's house through the front door\", \"the front door\" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to).\n\n[Discourse analysis](https:\/\/en.wikipedia.org\/wiki\/Discourse_analysis \"Discourse analysis\")\n\nThis rubric includes several related tasks. One task is discourse parsing, i.e., identifying the [discourse](https:\/\/en.wikipedia.org\/wiki\/Discourse \"Discourse\") structure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast). Another possible task is recognizing and classifying the [speech acts](https:\/\/en.wikipedia.org\/wiki\/Speech_act \"Speech act\") in a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.).\n\nImplicit semantic role labelling\n\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https:\/\/en.wikipedia.org\/wiki\/Frame_semantics_\\(linguistics\\) \"Frame semantics (linguistics)\")) and their explicit semantic roles in the current sentence (see [Semantic role labelling](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Semantic_role_labelling) above). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to [pro-drop languages](https:\/\/en.wikipedia.org\/wiki\/Pro-drop_language \"Pro-drop language\").\n\n[Recognizing textual entailment](https:\/\/en.wikipedia.org\/wiki\/Textual_entailment \"Textual entailment\")\n\nGiven two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false.[\\[33\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-rte:11-33)\n\n[Topic segmentation](https:\/\/en.wikipedia.org\/wiki\/Topic_segmentation \"Topic segmentation\") and recognition\n\nGiven a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment.\n\n[Argument mining](https:\/\/en.wikipedia.org\/wiki\/Argument_mining \"Argument mining\")\n\nThe goal of argument mining is the automatic extraction and identification of argumentative structures from [natural language](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") text with the aid of computer programs.[\\[34\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-34) Such argumentative structures include the premise, conclusions, the [argument scheme](https:\/\/en.wikipedia.org\/wiki\/Argument_scheme \"Argument scheme\") and the relationship between the main and subsidiary argument, or the main and counter-argument within discourse.[\\[35\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-35)[\\[36\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-36)\n\n### Higher-level NLP applications\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=14 \"Edit section: Higher-level NLP applications\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/4\/43\/Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png\/250px-Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)](https:\/\/en.wikipedia.org\/wiki\/File:Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)\n\nMachine translation in Firefox\n\n[Automatic summarization](https:\/\/en.wikipedia.org\/wiki\/Automatic_summarization \"Automatic summarization\") (text summarization)\n\nProduce a readable summary of a chunk of text. Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper.\n\nGrammatical error correction\n\nGrammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology\/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011.[\\[37\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-37)[\\[38\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-38)[\\[39\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-39) As far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as [GPT-2](https:\/\/en.wikipedia.org\/wiki\/GPT-2 \"GPT-2\"), this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications.\n\n[Logic translation](https:\/\/en.wikipedia.org\/wiki\/Logic_translation \"Logic translation\")\n\nTranslate a text from a natural language into formal logic.\n\n[Machine translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\") (MT)\n\nAutomatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed \"[AI-complete](https:\/\/en.wikipedia.org\/wiki\/AI-complete \"AI-complete\")\", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly.\n\n[Natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_understanding \"Natural language understanding\") (NLU)\n\nConvert chunks of text into more formal representations such as [first-order logic](https:\/\/en.wikipedia.org\/wiki\/First-order_logic \"First-order logic\") structures that are easier for [computer](https:\/\/en.wikipedia.org\/wiki\/Computer \"Computer\") programs to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as [closed-world assumption](https:\/\/en.wikipedia.org\/wiki\/Closed-world_assumption \"Closed-world assumption\") (CWA) vs. [open-world assumption](https:\/\/en.wikipedia.org\/wiki\/Open-world_assumption \"Open-world assumption\"), or subjective Yes\/No vs. objective True\/False is expected for the construction of a basis of semantics formalization.[\\[40\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-40)\n\n[Natural language generation](https:\/\/en.wikipedia.org\/wiki\/Natural_language_generation \"Natural language generation\") (NLG):\n\nConvert information from computer databases or semantic intents into readable human language.\n\nBook generation\n\nNot an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter, *The policeman's beard is half-constructed*).[\\[41\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-41) The first published work by a neural network was published in 2018, *[1 the Road](https:\/\/en.wikipedia.org\/wiki\/1_the_Road \"1 the Road\")*, marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free) [language models](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\"). The first machine-generated science book was published in 2019 (Beta Writer, *Lithium-Ion Batteries*, Springer, Cham).[\\[42\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-42) Unlike *Racter* and *1 the Road*, this is grounded on factual knowledge and based on text summarization.\n\n[Document AI](https:\/\/en.wikipedia.org\/wiki\/Document_AI \"Document AI\")\n\nA Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants.[\\[43\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-43)\n\n[Dialogue management](https:\/\/en.wikipedia.org\/wiki\/Dialogue_system \"Dialogue system\")\n\nComputer systems intended to converse with a human.\n\n[Question answering](https:\/\/en.wikipedia.org\/wiki\/Question_answering \"Question answering\")\n\nGiven a human-language question, determine its answer. Typical questions have a specific right answer (such as \"What is the capital of Canada?\"), but sometimes open-ended questions are also considered (such as \"What is the meaning of life?\").\n\n[Text-to-image generation](https:\/\/en.wikipedia.org\/wiki\/Text-to-image_generation \"Text-to-image generation\")\n\nGiven a description of an image, generate an image that matches the description.[\\[44\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-44)\n\nText-to-scene generation\n\nGiven a description of a scene, generate a [3D model](https:\/\/en.wikipedia.org\/wiki\/3D_model \"3D model\") of the scene.[\\[45\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-45)[\\[46\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-46)\n\n[Text-to-video](https:\/\/en.wikipedia.org\/wiki\/Text-to-video_model \"Text-to-video model\")\n\nGiven a description of a video, generate a video that matches the description.[\\[47\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-47)[\\[48\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-48)\n\n## General tendencies and (possible) future directions\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=15 \"Edit section: General tendencies and (possible) future directions\")\\]\n\nBased on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed:[\\[49\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-49)\n\n- Interest on increasingly abstract, \"cognitive\" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09\/2017–18: dependency syntax, 2004–05\/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing).\n- Increasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+\/100+ languages)\n- Elimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems)\n\n### Cognition\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=16 \"Edit section: Cognition\")\\]\n\nMost higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above).\n\n[Cognition](https:\/\/en.wikipedia.org\/wiki\/Cognition \"Cognition\") refers to \"the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses.\"[\\[50\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-50) [Cognitive science](https:\/\/en.wikipedia.org\/wiki\/Cognitive_science \"Cognitive science\") is the interdisciplinary, scientific study of the mind and its processes.[\\[51\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-51) [Cognitive linguistics](https:\/\/en.wikipedia.org\/wiki\/Cognitive_linguistics \"Cognitive linguistics\") is an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics.[\\[52\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-52) Especially during the age of [symbolic NLP](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Symbolic_NLP_\\(1950s_%E2%80%93_early_1990s\\)), the area of computational linguistics maintained strong ties with cognitive studies.\n\nAs an example, [George Lakoff](https:\/\/en.wikipedia.org\/wiki\/George_Lakoff \"George Lakoff\") offers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics,[\\[53\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-53) with two defining aspects:\n\n1. Apply the theory of [conceptual metaphor](https:\/\/en.wikipedia.org\/wiki\/Conceptual_metaphor \"Conceptual metaphor\"), explained by Lakoff as \"the understanding of one idea, in terms of another\" which provides an idea of the intent of the author.[\\[54\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-54) For example, consider the English word *big*. When used in a comparison (\"That is a big tree\"), the author's intent is to imply that the tree is *physically large* relative to other trees or the authors experience. When used metaphorically (\"Tomorrow is a big day\"), the author's intent to imply *importance*. The intent behind other usages, like in \"She is a big person\", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information.\n2. Assign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a [probabilistic context-free grammar](https:\/\/en.wikipedia.org\/wiki\/Probabilistic_context-free_grammar \"Probabilistic context-free grammar\") (PCFG). The mathematical equation for such algorithms is presented in [US Patent 9269353](https:\/\/worldwide.espacenet.com\/patent\/search\/family\/055314712\/publication\/US9269353B1?q=pn%3DUS9269353):[\\[55\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-55)\n\nR\n\nM\n\nM\n\n(\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n)\n\n\\=\n\nP\n\nM\n\nM\n\n(\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n)\n\n×\n\n1\n\n2\n\nd\n\n(\n\n∑\n\ni\n\n\\=\n\n−\n\nd\n\nd\n\n(\n\n(\n\nP\n\nM\n\nM\n\n(\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n)\n\n×\n\nP\n\nF\n\n(\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n−\n\ni\n\n,\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n,\n\nt\n\no\n\nk\n\ne\n\nn\n\nN\n\n\\+\n\ni\n\n)\n\n)\n\ni\n\n)\n\n{\\\\displaystyle {RMM(token\\_{N})}={PMM(token\\_{N})}\\\\times {\\\\frac {1}{2d}}\\\\left(\\\\sum \\_{i=-d}^{d}{((PMM(token\\_{N})}\\\\times {PF(token\\_{N-i},token\\_{N},token\\_{N+i}))\\_{i}}\\\\right)}\n\n![{\\\\displaystyle {RMM(token\\_{N})}={PMM(token\\_{N})}\\\\times {\\\\frac {1}{2d}}\\\\left(\\\\sum \\_{i=-d}^{d}{((PMM(token\\_{N})}\\\\times {PF(token\\_{N-i},token\\_{N},token\\_{N+i}))\\_{i}}\\\\right)}](https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/43ccadd794a4b84e20d1209997a463342e0dfbfe)\n\n*Where*\n\n**RMM** is the relative measure of meaning\n\n**token** is any block of text, sentence, phrase or word\n\n**N** is the number of tokens being analyzed\n\n**PMM** is the probable measure of meaning based on a corpora\n\n**d** is the non zero location of the token along the sequence of **N** tokens\n\n**PF** is the probability function specific to a language\n\nTies with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar,[\\[56\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-56) functional grammar,[\\[57\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-57) construction grammar,[\\[58\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-58) computational psycholinguistics and cognitive neuroscience (e.g., [ACT-R](https:\/\/en.wikipedia.org\/wiki\/ACT-R \"ACT-R\")), however, with limited uptake in mainstream NLP (as measured by presence on major conferences[\\[59\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-59) of the [ACL](https:\/\/en.wikipedia.org\/wiki\/Association_for_Computational_Linguistics \"Association for Computational Linguistics\")). More recently, ideas of cognitive NLP have been revived as an approach to achieve [explainability](https:\/\/en.wikipedia.org\/wiki\/Explainable_artificial_intelligence \"Explainable artificial intelligence\"), e.g., under the notion of \"cognitive AI\".[\\[60\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-60) Likewise, ideas of cognitive NLP are inherent to neural models [multimodal](https:\/\/en.wikipedia.org\/wiki\/Multimodal_interaction \"Multimodal interaction\") NLP (although rarely made explicit)[\\[61\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-61) and developments in [artificial intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence \"Artificial intelligence\"), specifically tools and technologies using [large language model](https:\/\/en.wikipedia.org\/wiki\/Large_language_model \"Large language model\") approaches[\\[62\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-62) and new directions in [artificial general intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_general_intelligence \"Artificial general intelligence\") based on the [free energy principle](https:\/\/en.wikipedia.org\/wiki\/Free_energy_principle \"Free energy principle\")[\\[63\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-63) by British neuroscientist and theoretician at University College London [Karl J. Friston](https:\/\/en.wikipedia.org\/wiki\/Karl_J._Friston \"Karl J. Friston\").\n\n## See also\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=17 \"Edit section: See also\")\\]\n\n- *[1 the Road](https:\/\/en.wikipedia.org\/wiki\/1_the_Road \"1 the Road\")*\n- [Artificial intelligence detection software](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_detection_software \"Artificial intelligence detection software\")\n- [Automated essay scoring](https:\/\/en.wikipedia.org\/wiki\/Automated_essay_scoring \"Automated essay scoring\")\n- [Biomedical text mining](https:\/\/en.wikipedia.org\/wiki\/Biomedical_text_mining \"Biomedical text mining\")\n- [Compound term processing](https:\/\/en.wikipedia.org\/wiki\/Compound_term_processing \"Compound term processing\")\n- [Computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\")\n- [Computer-assisted reviewing](https:\/\/en.wikipedia.org\/wiki\/Computer-assisted_reviewing \"Computer-assisted reviewing\")\n- [Controlled natural language](https:\/\/en.wikipedia.org\/wiki\/Controlled_natural_language \"Controlled natural language\")\n- [Deep learning](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\")\n- [Deep linguistic processing](https:\/\/en.wikipedia.org\/wiki\/Deep_linguistic_processing \"Deep linguistic processing\")\n- [Distributional semantics](https:\/\/en.wikipedia.org\/wiki\/Distributional_semantics \"Distributional semantics\")\n- [Foreign language reading aid](https:\/\/en.wikipedia.org\/wiki\/Foreign_language_reading_aid \"Foreign language reading aid\")\n- [Foreign language writing aid](https:\/\/en.wikipedia.org\/wiki\/Foreign_language_writing_aid \"Foreign language writing aid\")\n- [Information extraction](https:\/\/en.wikipedia.org\/wiki\/Information_extraction \"Information extraction\")\n- [Information retrieval](https:\/\/en.wikipedia.org\/wiki\/Information_retrieval \"Information retrieval\")\n- [Language and Communication Technologies](https:\/\/en.wikipedia.org\/wiki\/Language_and_Communication_Technologies \"Language and Communication Technologies\")\n- [Language model](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\")\n- [Language technology](https:\/\/en.wikipedia.org\/wiki\/Language_technology \"Language technology\")\n- [Latent semantic indexing](https:\/\/en.wikipedia.org\/wiki\/Latent_semantic_indexing \"Latent semantic indexing\")\n- [Multi-agent system](https:\/\/en.wikipedia.org\/wiki\/Multi-agent_system \"Multi-agent system\")\n- [Native-language identification](https:\/\/en.wikipedia.org\/wiki\/Native-language_identification \"Native-language identification\")\n- [Natural-language programming](https:\/\/en.wikipedia.org\/wiki\/Natural-language_programming \"Natural-language programming\")\n- [Natural-language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural-language_understanding \"Natural-language understanding\")\n- [Natural-language search](https:\/\/en.wikipedia.org\/wiki\/Natural-language_user_interface \"Natural-language user interface\")\n- [Outline of natural language processing](https:\/\/en.wikipedia.org\/wiki\/Outline_of_natural_language_processing \"Outline of natural language processing\")\n- [Query expansion](https:\/\/en.wikipedia.org\/wiki\/Query_expansion \"Query expansion\")\n- [Query understanding](https:\/\/en.wikipedia.org\/wiki\/Query_understanding \"Query understanding\")\n- [Reification (linguistics)](https:\/\/en.wikipedia.org\/wiki\/Reification_\\(linguistics\\) \"Reification (linguistics)\")\n- [Speech processing](https:\/\/en.wikipedia.org\/wiki\/Speech_processing \"Speech processing\")\n- [Spoken dialogue systems](https:\/\/en.wikipedia.org\/wiki\/Spoken_dialogue_systems \"Spoken dialogue systems\")\n- [Text-proofing](https:\/\/en.wikipedia.org\/wiki\/Text-proofing \"Text-proofing\")\n- [Text simplification](https:\/\/en.wikipedia.org\/wiki\/Text_simplification \"Text simplification\")\n- [Transformer (machine learning model)](https:\/\/en.wikipedia.org\/wiki\/Transformer_\\(machine_learning_model\\) \"Transformer (machine learning model)\")\n- [Truecasing](https:\/\/en.wikipedia.org\/wiki\/Truecasing \"Truecasing\")\n- [Question answering](https:\/\/en.wikipedia.org\/wiki\/Question_answering \"Question answering\")\n- [Word2vec](https:\/\/en.wikipedia.org\/wiki\/Word2vec \"Word2vec\")\n\n## References\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=18 \"Edit section: References\")\\]\n\n1. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-nlpintro_1-0)**\n   Eisenstein, Jacob (October 1, 2019). [*Introduction to Natural Language Processing*](https:\/\/mitpress.mit.edu\/9780262042840\/introduction-to-natural-language-processing\/). The MIT Press. p. 1. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n   \n   [978-0-262-04284-0](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-04284-0 \"Special:BookSources\/978-0-262-04284-0\")\n   \n   .\n2. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-2)**\n   [\"NLP\"](https:\/\/cs.stanford.edu\/people\/eroberts\/courses\/soco\/projects\/2004-05\/nlp\/overview_history.html).\n3. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-3)**\n   Hutchins, J. (2005). [\"The history of machine translation in a nutshell\"](https:\/\/web.archive.org\/web\/20190713103044\/http:\/\/www.hutchinsweb.me.uk\/Nutshell-2005.pdf) (PDF). Archived from [the original](http:\/\/www.hutchinsweb.me.uk\/Nutshell-2005.pdf) (PDF) on 2019-07-13. Retrieved 2019-02-04.\n   \n   \\[*[self-published source](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Verifiability#Self-published_sources \"Wikipedia:Verifiability\")*\\]\n4. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-4)** \"ALPAC: the (in)famous report\", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12.\n5. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-5)** [Crevier 1993](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#CITEREFCrevier1993), pp. 146–148 harvnb error: no target: CITEREFCrevier1993 ([help](https:\/\/en.wikipedia.org\/wiki\/Category:Harv_and_Sfn_template_errors \"Category:Harv and Sfn template errors\")), see also [Buchanan 2005](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#CITEREFBuchanan2005), p. 56 harvnb error: no target: CITEREFBuchanan2005 ([help](https:\/\/en.wikipedia.org\/wiki\/Category:Harv_and_Sfn_template_errors \"Category:Harv and Sfn template errors\")): \"Early programs were necessarily limited in scope by the size and speed of memory\"\n6. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-6)**\n   [Koskenniemi, Kimmo](https:\/\/en.wikipedia.org\/wiki\/Kimmo_Koskenniemi \"Kimmo Koskenniemi\") (1983), [*Two-level morphology: A general computational model of word-form recognition and production*](https:\/\/web.archive.org\/web\/20181221032913\/http:\/\/www.ling.helsinki.fi\/~koskenni\/doc\/Two-LevelMorphology.pdf) (PDF), Department of General Linguistics, [University of Helsinki](https:\/\/en.wikipedia.org\/wiki\/University_of_Helsinki \"University of Helsinki\"), archived from [the original](http:\/\/www.ling.helsinki.fi\/~koskenni\/doc\/Two-LevelMorphology.pdf) (PDF) on 2018-12-21, retrieved 2020-08-20\n7. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-7)** Joshi, A. K., & Weinstein, S. (1981, August). [Control of Inference: Role of Some Aspects of Discourse Structure-Centering](https:\/\/www.ijcai.org\/Proceedings\/81-1\/Papers\/071.pdf). In *IJCAI* (pp. 385–387).\n8. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-8)**\n   Guida, G.; Mauri, G. (July 1986). \"Evaluation of natural language processing systems: Issues and approaches\". *Proceedings of the IEEE*. **74** (7): 1026–1035\\. [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[1986IEEEP..74.1026G](https:\/\/ui.adsabs.harvard.edu\/abs\/1986IEEEP..74.1026G). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1109\/PROC.1986.13580](https:\/\/doi.org\/10.1109%2FPROC.1986.13580). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1558-2256](https:\/\/search.worldcat.org\/issn\/1558-2256). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [30688575](https:\/\/api.semanticscholar.org\/CorpusID:30688575).\n9. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-9)** Chomskyan linguistics encourages the investigation of \"[corner cases](https:\/\/en.wikipedia.org\/wiki\/Corner_case \"Corner case\")\" that stress the limits of its theoretical models (comparable to [pathological](https:\/\/en.wikipedia.org\/wiki\/Pathological_\\(mathematics\\) \"Pathological (mathematics)\") phenomena in mathematics), typically created using [thought experiments](https:\/\/en.wikipedia.org\/wiki\/Thought_experiment \"Thought experiment\"), rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in [corpus linguistics](https:\/\/en.wikipedia.org\/wiki\/Corpus_linguistics \"Corpus linguistics\"). The creation and use of such [corpora](https:\/\/en.wikipedia.org\/wiki\/Text_corpus \"Text corpus\") of real-world data is a fundamental part of machine-learning algorithms for natural language processing. In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called \"[poverty of the stimulus](https:\/\/en.wikipedia.org\/wiki\/Poverty_of_the_stimulus \"Poverty of the stimulus\")\" argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing. As a result, the Chomskyan paradigm discouraged the application of such models to language processing.\n10. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-10)**\n    Bengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003). [\"A neural probabilistic language model\"](https:\/\/dl.acm.org\/doi\/10.5555\/944919.944966). *The Journal of Machine Learning Research*. **3**: 1137–1155 – via ACM Digital Library.\n11. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-11)**\n    Mikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010). [\"Recurrent neural network based language model\"](https:\/\/gwern.net\/doc\/ai\/nn\/rnn\/2010-mikolov.pdf) (PDF). *Interspeech 2010*. pp. 1045–1048\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.21437\/Interspeech.2010-343](https:\/\/doi.org\/10.21437%2FInterspeech.2010-343). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [17048224](https:\/\/api.semanticscholar.org\/CorpusID:17048224).\n    \n    `{{cite book}}`: `|journal=` ignored ([help](https:\/\/en.wikipedia.org\/wiki\/Help:CS1_errors#periodical_ignored \"Help:CS1 errors\"))\n12. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-goldberg:nnlp17_12-0)**\n    Goldberg, Yoav (2016). [\"A Primer on Neural Network Models for Natural Language Processing\"](https:\/\/doi.org\/10.1613%2Fjair.4992). *Journal of Artificial Intelligence Research*. **57**: 345–420\\. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1807\\.10854](https:\/\/arxiv.org\/abs\/1807.10854). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1613\/jair.4992](https:\/\/doi.org\/10.1613%2Fjair.4992). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [8273530](https:\/\/api.semanticscholar.org\/CorpusID:8273530).\n13. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-goodfellow:book16_13-0)**\n    Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016). [*Deep Learning*](http:\/\/www.deeplearningbook.org\/). MIT Press.\n14. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-jozefowicz:lm16_14-0)**\n    Jozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016). *Exploring the Limits of Language Modeling*. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1602\\.02410](https:\/\/arxiv.org\/abs\/1602.02410). [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[2016arXiv160202410J](https:\/\/ui.adsabs.harvard.edu\/abs\/2016arXiv160202410J).\n15. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-choe:emnlp16_15-0)**\n    Choe, Do Kook; Charniak, Eugene. [\"Parsing as Language Modeling\"](https:\/\/web.archive.org\/web\/20181023034804\/https:\/\/aclanthology.coli.uni-saarland.de\/papers\/D16-1257\/d16-1257). *Emnlp 2016*. Archived from [the original](https:\/\/aclanthology.coli.uni-saarland.de\/papers\/D16-1257\/d16-1257) on 2018-10-23. Retrieved 2018-10-22.\n16. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-vinyals:nips15_16-0)**\n    Vinyals, Oriol; et al. (2014). [\"Grammar as a Foreign Language\"](https:\/\/papers.nips.cc\/paper\/5635-grammar-as-a-foreign-language.pdf) (PDF). *Nips2015*. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1412\\.7449](https:\/\/arxiv.org\/abs\/1412.7449). [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[2014arXiv1412.7449V](https:\/\/ui.adsabs.harvard.edu\/abs\/2014arXiv1412.7449V).\n17. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-17)**\n    Turchin, Alexander; Florez Builes, Luisa F. (2021-03-19). [\"Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC8120048). *Journal of Diabetes Science and Technology*. **15** (3): 553–560\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1177\/19322968211000831](https:\/\/doi.org\/10.1177%2F19322968211000831). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1932-2968](https:\/\/search.worldcat.org\/issn\/1932-2968). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [8120048](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC8120048). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [33736486](https:\/\/pubmed.ncbi.nlm.nih.gov\/33736486).\n18. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-18)**\n    Lee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10). [\"Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9233261). *JMIR Medical Informatics*. **10** (6) e38482. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.2196\/38482](https:\/\/doi.org\/10.2196%2F38482). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [2291-9694](https:\/\/search.worldcat.org\/issn\/2291-9694). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [9233261](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9233261). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [35687381](https:\/\/pubmed.ncbi.nlm.nih.gov\/35687381).\n19. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-winograd:shrdlu71_19-0)**\n    Winograd, Terry (1971). [*Procedures as a Representation for Data in a Computer Program for Understanding Natural Language*](http:\/\/hci.stanford.edu\/winograd\/shrdlu\/) (Thesis).\n20. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-schank77_20-0)**\n    Schank, Roger C.; Abelson, Robert P. (1977). *Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures*. Hillsdale: Erlbaum. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [0-470-99033-3](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/0-470-99033-3 \"Special:BookSources\/0-470-99033-3\")\n    \n    .\n21. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-johnson:eacl:ilcl09_21-0)** [Mark Johnson. How the statistical revolution changes (computational) linguistics.](http:\/\/www.aclweb.org\/anthology\/W09-0103) Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics.\n22. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-resnik:langlog11_22-0)** [Philip Resnik. Four revolutions.](http:\/\/languagelog.ldc.upenn.edu\/nll\/?p=2946) Language Log, February 5, 2011.\n23. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-23)**\n    Socher, Richard. [\"Deep Learning For NLP-ACL 2012 Tutorial\"](https:\/\/web.archive.org\/web\/20210414054126\/https:\/\/www.socher.org\/index.php\/Main\/DeepLearningForNLP-ACL2012Tutorial). *www.socher.org*. Archived from [the original](https:\/\/www.socher.org\/index.php\/Main\/DeepLearningForNLP-ACL2012Tutorial) on 2021-04-14. Retrieved 2020-08-17.\n    \n    This was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. \\[Link is broken, try <http:\/\/web.stanford.edu\/class\/cs224n\/>\\]\n24. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-24)**\n    Segev, Elad (2022). [*Semantic Network Analysis in Social Sciences*](https:\/\/www.routledge.com\/Semantic-Network-Analysis-in-Social-Sciences\/Segev\/p\/book\/9780367636524). London: Routledge. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-367-63652-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-367-63652-4 \"Special:BookSources\/978-0-367-63652-4\")\n    \n    . [Archived](https:\/\/web.archive.org\/web\/20211205140726\/https:\/\/www.routledge.com\/Semantic-Network-Analysis-in-Social-Sciences\/Segev\/p\/book\/9780367636524) from the original on 5 December 2021. Retrieved 5 December 2021.\n25. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-25)**\n    Yi, Chucai; [Tian, Yingli](https:\/\/en.wikipedia.org\/wiki\/Yingli_Tian \"Yingli Tian\") (2012), \"Assistive Text Reading from Complex Background for Blind Persons\", *Camera-Based Document Analysis and Recognition*, Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. 15–28, [CiteSeerX](https:\/\/en.wikipedia.org\/wiki\/CiteSeerX_\\(identifier\\) \"CiteSeerX (identifier)\") [10\\.1.1.668.869](https:\/\/citeseerx.ist.psu.edu\/viewdoc\/summary?doi=10.1.1.668.869), [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1007\/978-3-642-29364-1\\_2](https:\/\/doi.org\/10.1007%2F978-3-642-29364-1_2), [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-3-642-29363-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-3-642-29363-4 \"Special:BookSources\/978-3-642-29363-4\")\n    \n    `{{citation}}`: CS1 maint: work parameter with ISBN ([link](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_work_parameter_with_ISBN \"Category:CS1 maint: work parameter with ISBN\"))\n26. ^ [***a***](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-:0_26-0) [***b***](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-:0_26-1)\n    [\"Natural Language Processing (NLP) - A Complete Guide\"](https:\/\/www.deeplearning.ai\/resources\/natural-language-processing\/). *www.deeplearning.ai*. 2023-01-11. Retrieved 2024-05-05.\n27. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-27)**\n    [\"GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks\"](https:\/\/www.geeksforgeeks.org\/nlp\/tokenization-in-natural-language-processing-nlp\/). *Geeksforgeeks*. 25 June 2024.\n28. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-28)**\n    Torabi, Mohhamadreza; Ghasemi, Fahimeh. [\"Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC12391535). *Scientific Reports*. **1** (1): 1–15\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1038\/s41598-025-14699-1](https:\/\/doi.org\/10.1038%2Fs41598-025-14699-1). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [12391535](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC12391535).\n29. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-29)**\n    [\"What is Natural Language Processing? Intro to NLP in Machine Learning\"](https:\/\/www.gyansetu.in\/what-is-natural-language-processing\/). *GyanSetu\\!*. 2020-12-06. Retrieved 2021-01-09.\n30. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-30)**\n    Kishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012). [\"Manipuri Morpheme Identification\"](http:\/\/aclweb.org\/anthology\/\/W\/W12\/W12-5008.pdf) (PDF). *Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP)*. COLING 2012, Mumbai, December 2012: 95–108\\.\n    \n    `{{cite journal}}`: CS1 maint: location ([link](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_location \"Category:CS1 maint: location\"))\n31. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-31)**\n    Klein, Dan; Manning, Christopher D. (2002). [\"Natural language grammar induction using a constituent-context model\"](http:\/\/papers.nips.cc\/paper\/1945-natural-language-grammar-induction-using-a-constituent-context-model.pdf) (PDF). *Advances in Neural Information Processing Systems*.\n32. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-32)**\n    Kariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023). [\"Precision information extraction for rare disease epidemiology at scale\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9972634). *Journal of Translational Medicine*. **21** (1): 157. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1186\/s12967-023-04011-y](https:\/\/doi.org\/10.1186%2Fs12967-023-04011-y). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [9972634](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9972634). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [36855134](https:\/\/pubmed.ncbi.nlm.nih.gov\/36855134).\n33. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-rte:11_33-0)** PASCAL Recognizing Textual Entailment Challenge (RTE-7) <https:\/\/tac.nist.gov\/\/2011\/RTE\/>\n34. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-34)**\n    Lippi, Marco; Torroni, Paolo (2016-04-20). [\"Argumentation Mining: State of the Art and Emerging Trends\"](https:\/\/dl.acm.org\/doi\/10.1145\/2850417). *ACM Transactions on Internet Technology*. **16** (2): 1–25\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1145\/2850417](https:\/\/doi.org\/10.1145%2F2850417). [hdl](https:\/\/en.wikipedia.org\/wiki\/Hdl_\\(identifier\\) \"Hdl (identifier)\"):[11585\/523460](https:\/\/hdl.handle.net\/11585%2F523460). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1533-5399](https:\/\/search.worldcat.org\/issn\/1533-5399). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [9561587](https:\/\/api.semanticscholar.org\/CorpusID:9561587).\n35. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-35)**\n    [\"Argument Mining – IJCAI2016 Tutorial\"](https:\/\/web.archive.org\/web\/20210418083659\/https:\/\/www.i3s.unice.fr\/~villata\/tutorialIJCAI2016.html). *www.i3s.unice.fr*. Archived from [the original](https:\/\/www.i3s.unice.fr\/~villata\/tutorialIJCAI2016.html) on 2021-04-18. Retrieved 2021-03-09.\n36. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-36)**\n    [\"NLP Approaches to Computational Argumentation – ACL 2016, Berlin\"](http:\/\/acl2016tutorial.arg.tech\/). Retrieved 2021-03-09.\n37. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-37)**\n    Administration. [\"Centre for Language Technology (CLT)\"](https:\/\/www.mq.edu.au\/research\/research-centres-groups-and-facilities\/innovative-technologies\/centres\/centre-for-language-technology-clt). *Macquarie University*. Retrieved 2021-01-11.\n38. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-38)**\n    [\"Shared Task: Grammatical Error Correction\"](https:\/\/www.comp.nus.edu.sg\/~nlp\/conll13st.html). *www.comp.nus.edu.sg*. Retrieved 2021-01-11.\n39. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-39)**\n    [\"Shared Task: Grammatical Error Correction\"](https:\/\/www.comp.nus.edu.sg\/~nlp\/conll14st.html). *www.comp.nus.edu.sg*. Retrieved 2021-01-11.\n40. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-40)**\n    Duan, Yucong; Cruz, Christophe (2011). [\"Formalizing Semantic of Natural Language through Conceptualization from Existence\"](https:\/\/web.archive.org\/web\/20111009135952\/http:\/\/www.ijimt.org\/abstract\/100-E00187.htm). *International Journal of Innovation, Management and Technology*. **2** (1): 37–42\\. Archived from [the original](http:\/\/www.ijimt.org\/abstract\/100-E00187.htm) on 2011-10-09.\n41. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-41)**\n    [\"U B U W E B :: Racter\"](http:\/\/www.ubu.com\/historical\/racter\/index.html). *www.ubu.com*. Retrieved 2020-08-17.\n42. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-42)**\n    Writer, Beta (2019). *Lithium-Ion Batteries*. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1007\/978-3-030-16800-1](https:\/\/doi.org\/10.1007%2F978-3-030-16800-1). [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-3-030-16799-8](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-3-030-16799-8 \"Special:BookSources\/978-3-030-16799-8\")\n    \n    . [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [155818532](https:\/\/api.semanticscholar.org\/CorpusID:155818532).\n43. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-43)**\n    [\"Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube\"](https:\/\/ghostarchive.org\/varchive\/youtube\/20211030\/7dtl650D0y0). *www.youtube.com*. 11 April 2019. Archived from [the original](https:\/\/www.youtube.com\/watch?v=7dtl650D0y0) on 2021-10-30. Retrieved 2021-01-11.\n44. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-44)**\n    Robertson, Adi (2022-04-06). [\"OpenAI's DALL-E AI image generator can now edit pictures, too\"](https:\/\/www.theverge.com\/2022\/4\/6\/23012123\/openai-clip-dalle-2-ai-text-to-image-generator-testing). *The Verge*. Retrieved 2022-06-07.\n45. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-45)**\n    [\"The Stanford Natural Language Processing Group\"](https:\/\/nlp.stanford.edu\/projects\/text2scene.shtml). *nlp.stanford.edu*. Retrieved 2022-06-07.\n46. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-46)**\n    Coyne, Bob; Sproat, Richard (2001-08-01). [\"WordsEye\"](https:\/\/doi.org\/10.1145\/383259.383316). *Proceedings of the 28th annual conference on Computer graphics and interactive techniques*. SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. 487–496\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1145\/383259.383316](https:\/\/doi.org\/10.1145%2F383259.383316). [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-1-58113-374-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1-58113-374-5 \"Special:BookSources\/978-1-58113-374-5\")\n    \n    . [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [3842372](https:\/\/api.semanticscholar.org\/CorpusID:3842372).\n47. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-47)**\n    [\"Google announces AI advances in text-to-video, language translation, more\"](https:\/\/venturebeat.com\/ai\/google-announces-ai-advances-in-text-to-video-language-translation-more\/). *VentureBeat*. 2022-11-02. Retrieved 2022-11-09.\n48. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-48)**\n    Vincent, James (2022-09-29). [\"Meta's new text-to-video AI generator is like DALL-E for video\"](https:\/\/www.theverge.com\/2022\/9\/29\/23378210\/meta-text-to-video-ai-generation-make-a-video-model-dall-e). *The Verge*. Retrieved 2022-11-09.\n49. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-49)**\n    [\"Previous shared tasks \\| CoNLL\"](https:\/\/www.conll.org\/previous-tasks). *www.conll.org*. Retrieved 2021-01-11.\n50. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-50)**\n    [\"Cognition\"](https:\/\/web.archive.org\/web\/20200715113427\/https:\/\/www.lexico.com\/definition\/cognition). *Lexico*. [Oxford University Press](https:\/\/en.wikipedia.org\/wiki\/Oxford_University_Press \"Oxford University Press\") and [Dictionary.com](https:\/\/en.wikipedia.org\/wiki\/Dictionary.com \"Dictionary.com\"). Archived from [the original](https:\/\/www.lexico.com\/definition\/cognition) on July 15, 2020. Retrieved 6 May 2020.\n51. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-51)**\n    [\"Ask the Cognitive Scientist\"](http:\/\/www.aft.org\/newspubs\/periodicals\/ae\/summer2002\/willingham.cfm). *American Federation of Teachers*. 8 August 2014. \"Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind.\"\n52. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-52)**\n    Robinson, Peter (2008). *Handbook of Cognitive Linguistics and Second Language Acquisition*. Routledge. pp. 3–8\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-805-85352-0](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-805-85352-0 \"Special:BookSources\/978-0-805-85352-0\")\n    \n    .\n53. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-53)**\n    Lakoff, George (1999). *Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm*. New York Basic Books. pp. 569–583\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-465-05674-3](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-465-05674-3 \"Special:BookSources\/978-0-465-05674-3\")\n    \n    .\n54. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-54)**\n    Strauss, Claudia (1999). *A Cognitive Theory of Cultural Meaning*. Cambridge University Press. pp. 156–164\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-521-59541-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-521-59541-4 \"Special:BookSources\/978-0-521-59541-4\")\n    \n    .\n55. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-55)** [US patent 9269353](https:\/\/worldwide.espacenet.com\/textdoc?DB=EPODOC&IDX=US9269353)\n56. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-56)**\n    [\"Universal Conceptual Cognitive Annotation (UCCA)\"](https:\/\/universalconceptualcognitiveannotation.github.io\/). *Universal Conceptual Cognitive Annotation (UCCA)*. Retrieved 2021-01-11.\n57. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-57)** Rodríguez, F. C., & Mairal-Usón, R. (2016). [Building an RRG computational grammar](https:\/\/www.redalyc.org\/pdf\/1345\/134549291020.pdf). *Onomazein*, (34), 86–117.\n58. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-58)**\n    [\"Fluid Construction Grammar – A fully operational processing system for construction grammars\"](https:\/\/www.fcg-net.org\/). Retrieved 2021-01-11.\n59. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-59)**\n    [\"ACL Member Portal \\| The Association for Computational Linguistics Member Portal\"](https:\/\/www.aclweb.org\/portal\/). *www.aclweb.org*. Retrieved 2021-01-11.\n60. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-60)**\n    [\"Chunks and Rules\"](https:\/\/www.w3.org\/Data\/demos\/chunks\/chunks.html). *W3C*. Retrieved 2021-01-11.\n61. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-61)**\n    Socher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014). [\"Grounded Compositional Semantics for Finding and Describing Images with Sentences\"](https:\/\/doi.org\/10.1162%2Ftacl_a_00177). *Transactions of the Association for Computational Linguistics*. **2**: 207–218\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1162\/tacl\\_a\\_00177](https:\/\/doi.org\/10.1162%2Ftacl_a_00177). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [2317858](https:\/\/api.semanticscholar.org\/CorpusID:2317858).\n62. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-62)**\n    Dasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). \"Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al\". [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[2207\\.07051](https:\/\/arxiv.org\/abs\/2207.07051) \\[[cs.CL](https:\/\/arxiv.org\/archive\/cs.CL)\\].\n63. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-63)**\n    Friston, Karl J. (2022). *Active Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference*. The MIT Press. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-262-36997-8](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-36997-8 \"Special:BookSources\/978-0-262-36997-8\")\n    \n    .\n\n## Further reading\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=19 \"Edit section: Further reading\")\\]\n\n- Bates, M (1995). [\"Models of natural language understanding\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC40721). *Proceedings of the National Academy of Sciences of the United States of America*. **92** (22): 9977–9982\\. [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[1995PNAS...92.9977B](https:\/\/ui.adsabs.harvard.edu\/abs\/1995PNAS...92.9977B). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1073\/pnas.92.22.9977](https:\/\/doi.org\/10.1073%2Fpnas.92.22.9977). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [40721](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC40721). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [7479812](https:\/\/pubmed.ncbi.nlm.nih.gov\/7479812).\n- Steven Bird, Ewan Klein, and Edward Loper (2009). *Natural Language Processing with Python*. O'Reilly Media.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-596-51649-9](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-596-51649-9 \"Special:BookSources\/978-0-596-51649-9\")\n  .\n- [Kenna Hughes-Castleberry](https:\/\/en.wikipedia.org\/w\/index.php?title=Kenna_Hughes-Castleberry&action=edit&redlink=1 \"Kenna Hughes-Castleberry (page does not exist)\"), \"A Murder Mystery Puzzle: The literary puzzle *[Cain's Jawbone](https:\/\/en.wikipedia.org\/wiki\/Cain%27s_Jawbone \"Cain's Jawbone\")*, which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms\", *[Scientific American](https:\/\/en.wikipedia.org\/wiki\/Scientific_American \"Scientific American\")*, vol. 329, no. 4 (November 2023), pp. 81–82. \"This murder mystery competition has revealed that although NLP ([natural-language processing](https:\/\/en.wikipedia.org\/wiki\/Natural-language_processing \"Natural-language processing\")) models are capable of incredible feats, their abilities are very much limited by the amount of [context](https:\/\/en.wikipedia.org\/wiki\/Context_\\(linguistics\\) \"Context (linguistics)\") they receive. This \\[...\\] could cause \\[difficulties\\] for researchers who hope to use them to do things such as analyze [ancient languages](https:\/\/en.wikipedia.org\/wiki\/Ancient_language \"Ancient language\"). In some cases, there are few historical records on long-gone [civilizations](https:\/\/en.wikipedia.org\/wiki\/Civilization \"Civilization\") to serve as [training data](https:\/\/en.wikipedia.org\/wiki\/Training_data \"Training data\") for such a purpose.\" (p. 82.)\n- Daniel Jurafsky and James H. Martin (2008). *Speech and Language Processing*, 2nd edition. Pearson Prentice Hall.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-13-187321-6](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-13-187321-6 \"Special:BookSources\/978-0-13-187321-6\")\n  .\n- Mohamed Zakaria Kurdi (2016). *Natural Language Processing and Computational Linguistics: speech, morphology, and syntax*, Volume 1. ISTE-Wiley.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-1848218482](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1848218482 \"Special:BookSources\/978-1848218482\")\n  .\n- Mohamed Zakaria Kurdi (2017). *Natural Language Processing and Computational Linguistics: semantics, discourse, and applications*, Volume 2. ISTE-Wiley.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-1848219212](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1848219212 \"Special:BookSources\/978-1848219212\")\n  .\n- Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008). *Introduction to Information Retrieval*. Cambridge University Press.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-521-86571-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-521-86571-5 \"Special:BookSources\/978-0-521-86571-5\")\n  . [Official html and pdf versions available without charge.](http:\/\/nlp.stanford.edu\/IR-book\/)\n- Christopher D. Manning and Hinrich Schütze (1999). *Foundations of Statistical Natural Language Processing*. The MIT Press.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-262-13360-9](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-13360-9 \"Special:BookSources\/978-0-262-13360-9\")\n  .\n- David M. W. Powers and Christopher C. R. Turk (1989). *Machine Learning of Natural Language*. Springer-Verlag.\n  [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-387-19557-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-387-19557-5 \"Special:BookSources\/978-0-387-19557-5\")\n  .\n\n## External links\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=20 \"Edit section: External links\")\\]\n\n- [![Wikimedia Commons logo](https:\/\/upload.wikimedia.org\/wikipedia\/en\/thumb\/4\/4a\/Commons-logo.svg\/20px-Commons-logo.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:Commons-logo.svg) Media related to [Natural language processing](https:\/\/commons.wikimedia.org\/wiki\/Category:Natural_language_processing \"commons:Category:Natural language processing\") at Wikimedia Commons\n\n| [v](https:\/\/en.wikipedia.org\/wiki\/Template:Natural_language_processing \"Template:Natural language processing\") [t](https:\/\/en.wikipedia.org\/wiki\/Template_talk:Natural_language_processing \"Template talk:Natural language processing\") [e](https:\/\/en.wikipedia.org\/wiki\/Special:EditPage\/Template:Natural_language_processing \"Special:EditPage\/Template:Natural language processing\")[Natural language processing]() |   |\n|---|---|\n| General terms | [AI-complete](https:\/\/en.wikipedia.org\/wiki\/AI-complete \"AI-complete\") [Bag-of-words](https:\/\/en.wikipedia.org\/wiki\/Bag-of-words_model \"Bag-of-words model\") [*n*\\-gram](https:\/\/en.wikipedia.org\/wiki\/N-gram \"N-gram\") [Bigram](https:\/\/en.wikipedia.org\/wiki\/Bigram \"Bigram\") [Trigram](https:\/\/en.wikipedia.org\/wiki\/Trigram \"Trigram\") [Computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\") [Natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_understanding \"Natural language understanding\") [Stop words](https:\/\/en.wikipedia.org\/wiki\/Stop_word \"Stop word\") [Text processing](https:\/\/en.wikipedia.org\/wiki\/Text_processing \"Text processing\") |\n| [Text analysis](https:\/\/en.wikipedia.org\/wiki\/Text_mining \"Text mining\") |   |\n|   |   |\n| [Text segmentation](https:\/\/en.wikipedia.org\/wiki\/Text_segmentation \"Text segmentation\") | [Compound-term processing](https:\/\/en.wikipedia.org\/wiki\/Compound-term_processing \"Compound-term processing\") [Lemmatization](https:\/\/en.wikipedia.org\/wiki\/Lemmatization \"Lemmatization\") [Lexical analysis](https:\/\/en.wikipedia.org\/wiki\/Lexical_analysis \"Lexical analysis\") [Text chunking](https:\/\/en.wikipedia.org\/wiki\/Shallow_parsing \"Shallow parsing\") [Stemming](https:\/\/en.wikipedia.org\/wiki\/Stemming \"Stemming\") [Sentence segmentation](https:\/\/en.wikipedia.org\/wiki\/Sentence_boundary_disambiguation \"Sentence boundary disambiguation\") [Word segmentation](https:\/\/en.wikipedia.org\/wiki\/Word#Word_boundaries \"Word\") |\n| [Automatic summarization](https:\/\/en.wikipedia.org\/wiki\/Automatic_summarization \"Automatic summarization\") | [Multi-document summarization](https:\/\/en.wikipedia.org\/wiki\/Multi-document_summarization \"Multi-document summarization\") [Sentence extraction](https:\/\/en.wikipedia.org\/wiki\/Sentence_extraction \"Sentence extraction\") [Text simplification](https:\/\/en.wikipedia.org\/wiki\/Text_simplification \"Text simplification\") |\n| [Machine translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\") | [Computer-assisted](https:\/\/en.wikipedia.org\/wiki\/Computer-assisted_translation \"Computer-assisted translation\") [Example-based](https:\/\/en.wikipedia.org\/wiki\/Example-based_machine_translation \"Example-based machine translation\") [Rule-based](https:\/\/en.wikipedia.org\/wiki\/Rule-based_machine_translation \"Rule-based machine translation\") [Statistical](https:\/\/en.wikipedia.org\/wiki\/Statistical_machine_translation \"Statistical machine translation\") [Transfer-based](https:\/\/en.wikipedia.org\/wiki\/Transfer-based_machine_translation \"Transfer-based machine translation\") [Neural](https:\/\/en.wikipedia.org\/wiki\/Neural_machine_translation \"Neural machine translation\") |\n| [Distributional semantics](https:\/\/en.wikipedia.org\/wiki\/Distributional_semantics \"Distributional semantics\") models | [BERT](https:\/\/en.wikipedia.org\/wiki\/BERT_\\(language_model\\) \"BERT (language model)\") [Document-term matrix](https:\/\/en.wikipedia.org\/wiki\/Document-term_matrix \"Document-term matrix\") [Explicit semantic analysis](https:\/\/en.wikipedia.org\/wiki\/Explicit_semantic_analysis \"Explicit semantic analysis\") [fastText](https:\/\/en.wikipedia.org\/wiki\/FastText \"FastText\") [GloVe](https:\/\/en.wikipedia.org\/wiki\/GloVe \"GloVe\") [Language model](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\") [large](https:\/\/en.wikipedia.org\/wiki\/Large_language_model \"Large language model\") [small](https:\/\/en.wikipedia.org\/wiki\/Small_language_model \"Small language model\") [Latent semantic analysis](https:\/\/en.wikipedia.org\/wiki\/Latent_semantic_analysis \"Latent semantic analysis\") [Long short-term memory](https:\/\/en.wikipedia.org\/wiki\/Long_short-term_memory \"Long short-term memory\") [Seq2seq](https:\/\/en.wikipedia.org\/wiki\/Seq2seq \"Seq2seq\") [Transformer](https:\/\/en.wikipedia.org\/wiki\/Transformer_\\(deep_learning_architecture\\) \"Transformer (deep learning architecture)\") [Word embedding](https:\/\/en.wikipedia.org\/wiki\/Word_embedding \"Word embedding\") [Word2vec](https:\/\/en.wikipedia.org\/wiki\/Word2vec \"Word2vec\") |\n| [Language resources](https:\/\/en.wikipedia.org\/wiki\/Language_resource \"Language resource\"), datasets and corpora |   |\n|   |   |\n| Types and standards | [Corpus linguistics](https:\/\/en.wikipedia.org\/wiki\/Corpus_linguistics \"Corpus linguistics\") [Lexical resource](https:\/\/en.wikipedia.org\/wiki\/Lexical_resource \"Lexical resource\") [Linguistic Linked Open Data](https:\/\/en.wikipedia.org\/wiki\/Linguistic_Linked_Open_Data \"Linguistic Linked Open Data\") [Machine-readable dictionary](https:\/\/en.wikipedia.org\/wiki\/Machine-readable_dictionary \"Machine-readable dictionary\") [Parallel text](https:\/\/en.wikipedia.org\/wiki\/Parallel_text \"Parallel text\") [PropBank](https:\/\/en.wikipedia.org\/wiki\/PropBank \"PropBank\") [Semantic network](https:\/\/en.wikipedia.org\/wiki\/Semantic_network \"Semantic network\") [Simple Knowledge Organization System](https:\/\/en.wikipedia.org\/wiki\/Simple_Knowledge_Organization_System \"Simple Knowledge Organization System\") [Speech corpus](https:\/\/en.wikipedia.org\/wiki\/Speech_corpus \"Speech corpus\") [Text corpus](https:\/\/en.wikipedia.org\/wiki\/Text_corpus \"Text corpus\") [Thesaurus (information retrieval)](https:\/\/en.wikipedia.org\/wiki\/Thesaurus_\\(information_retrieval\\) \"Thesaurus (information retrieval)\") [Treebank](https:\/\/en.wikipedia.org\/wiki\/Treebank \"Treebank\") [Universal Dependencies](https:\/\/en.wikipedia.org\/wiki\/Universal_Dependencies \"Universal Dependencies\") |\n| Data | [BabelNet](https:\/\/en.wikipedia.org\/wiki\/BabelNet \"BabelNet\") [Bank of English](https:\/\/en.wikipedia.org\/wiki\/Bank_of_English \"Bank of English\") [DBpedia](https:\/\/en.wikipedia.org\/wiki\/DBpedia \"DBpedia\") [FrameNet](https:\/\/en.wikipedia.org\/wiki\/FrameNet \"FrameNet\") [Google Ngram Viewer](https:\/\/en.wikipedia.org\/wiki\/Google_Ngram_Viewer \"Google Ngram Viewer\") [UBY](https:\/\/en.wikipedia.org\/wiki\/UBY \"UBY\") [WordNet](https:\/\/en.wikipedia.org\/wiki\/WordNet \"WordNet\") [Wikidata](https:\/\/en.wikipedia.org\/wiki\/Wikidata \"Wikidata\") |\n| [Automatic identification and data capture](https:\/\/en.wikipedia.org\/wiki\/Automatic_identification_and_data_capture \"Automatic identification and data capture\") | [Speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\") [Speech segmentation](https:\/\/en.wikipedia.org\/wiki\/Speech_segmentation \"Speech segmentation\") [Speech synthesis](https:\/\/en.wikipedia.org\/wiki\/Speech_synthesis \"Speech synthesis\") [Natural language generation](https:\/\/en.wikipedia.org\/wiki\/Natural_language_generation \"Natural language generation\") [Optical character recognition](https:\/\/en.wikipedia.org\/wiki\/Optical_character_recognition \"Optical character recognition\") |\n| [Topic model](https:\/\/en.wikipedia.org\/wiki\/Topic_model \"Topic model\") | [Document classification](https:\/\/en.wikipedia.org\/wiki\/Document_classification \"Document classification\") [Latent Dirichlet allocation](https:\/\/en.wikipedia.org\/wiki\/Latent_Dirichlet_allocation \"Latent Dirichlet allocation\") [Pachinko allocation](https:\/\/en.wikipedia.org\/wiki\/Pachinko_allocation \"Pachinko allocation\") |\n| [Computer-assisted reviewing](https:\/\/en.wikipedia.org\/wiki\/Computer-assisted_reviewing \"Computer-assisted reviewing\") | [Automated essay scoring](https:\/\/en.wikipedia.org\/wiki\/Automated_essay_scoring \"Automated essay scoring\") [Concordancer](https:\/\/en.wikipedia.org\/wiki\/Concordancer \"Concordancer\") [Grammar checker](https:\/\/en.wikipedia.org\/wiki\/Grammar_checker \"Grammar checker\") [Predictive text](https:\/\/en.wikipedia.org\/wiki\/Predictive_text \"Predictive text\") [Pronunciation assessment](https:\/\/en.wikipedia.org\/wiki\/Pronunciation_assessment \"Pronunciation assessment\") [Spell checker](https:\/\/en.wikipedia.org\/wiki\/Spell_checker \"Spell checker\") |\n| [Natural language user interface](https:\/\/en.wikipedia.org\/wiki\/Natural-language_user_interface \"Natural-language user interface\") | [Chatbot](https:\/\/en.wikipedia.org\/wiki\/Chatbot \"Chatbot\") [Interactive fiction](https:\/\/en.wikipedia.org\/wiki\/Interactive_fiction \"Interactive fiction\") [Question answering](https:\/\/en.wikipedia.org\/wiki\/Question_answering \"Question answering\") [Virtual assistant](https:\/\/en.wikipedia.org\/wiki\/Virtual_assistant \"Virtual assistant\") [Voice user interface](https:\/\/en.wikipedia.org\/wiki\/Voice_user_interface \"Voice user interface\") |\n| Related | [Formal semantics](https:\/\/en.wikipedia.org\/wiki\/Formal_semantics_\\(natural_language\\) \"Formal semantics (natural language)\") [Gensim](https:\/\/en.wikipedia.org\/wiki\/Gensim \"Gensim\") [Hallucination](https:\/\/en.wikipedia.org\/wiki\/Hallucination_\\(artificial_intelligence\\) \"Hallucination (artificial intelligence)\") [Natural Language Toolkit](https:\/\/en.wikipedia.org\/wiki\/Natural_Language_Toolkit \"Natural Language Toolkit\") [spaCy](https:\/\/en.wikipedia.org\/wiki\/SpaCy \"SpaCy\") |\n\n[Portal](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Contents\/Portals \"Wikipedia:Contents\/Portals\"):\n\n- [![icon](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/d\/de\/Globe_of_letters.svg\/20px-Globe_of_letters.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:Globe_of_letters.svg) [Language](https:\/\/en.wikipedia.org\/wiki\/Portal:Language \"Portal:Language\")\n\n| [Authority control databases](https:\/\/en.wikipedia.org\/wiki\/Help:Authority_control \"Help:Authority control\") [![Edit this at Wikidata](https:\/\/upload.wikimedia.org\/wikipedia\/en\/thumb\/8\/8a\/OOjs_UI_icon_edit-ltr-progressive.svg\/20px-OOjs_UI_icon_edit-ltr-progressive.svg.png)](https:\/\/www.wikidata.org\/wiki\/Q30642#identifiers \"Edit this at Wikidata\") |   |\n|---|---|\n| National | [United States](https:\/\/id.loc.gov\/authorities\/sh88002425) [Japan](https:\/\/id.ndl.go.jp\/auth\/ndlna\/00562347) [Czech Republic](https:\/\/aleph.nkp.cz\/F\/?func=find-c&local_base=aut&ccl_term=ica=ph427562&CON_LNG=ENG) [Israel](https:\/\/www.nli.org.il\/en\/authorities\/987007536703305171) |\n| Other | [Yale LUX](https:\/\/lux.collections.yale.edu\/view\/concept\/e0fc4a1c-7602-43f9-9cea-0d8bb49039b8) |\n\n![](https:\/\/en.wikipedia.org\/wiki\/Special:CentralAutoLogin\/start?useformat=desktop&type=1x1&usesul3=1)\n\nRetrieved from \"<https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&oldid=1343173705>\"\n\n[Categories](https:\/\/en.wikipedia.org\/wiki\/Help:Category \"Help:Category\"):\n\n- [Natural language processing](https:\/\/en.wikipedia.org\/wiki\/Category:Natural_language_processing \"Category:Natural language processing\")\n- [Computational fields of study](https:\/\/en.wikipedia.org\/wiki\/Category:Computational_fields_of_study \"Category:Computational fields of study\")\n- [Computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Category:Computational_linguistics \"Category:Computational linguistics\")\n- [Speech recognition](https:\/\/en.wikipedia.org\/wiki\/Category:Speech_recognition \"Category:Speech recognition\")\n\nHidden categories:\n\n- [All accuracy disputes](https:\/\/en.wikipedia.org\/wiki\/Category:All_accuracy_disputes \"Category:All accuracy disputes\")\n- [Accuracy disputes from December 2013](https:\/\/en.wikipedia.org\/wiki\/Category:Accuracy_disputes_from_December_2013 \"Category:Accuracy disputes from December 2013\")\n- [Harv and Sfn no-target errors](https:\/\/en.wikipedia.org\/wiki\/Category:Harv_and_Sfn_no-target_errors \"Category:Harv and Sfn no-target errors\")\n- [CS1 errors: periodical ignored](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_errors:_periodical_ignored \"Category:CS1 errors: periodical ignored\")\n- [CS1 maint: work parameter with ISBN](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_work_parameter_with_ISBN \"Category:CS1 maint: work parameter with ISBN\")\n- [CS1 maint: location](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_location \"Category:CS1 maint: location\")\n- [Articles with short description](https:\/\/en.wikipedia.org\/wiki\/Category:Articles_with_short_description \"Category:Articles with short description\")\n- [Short description is different from Wikidata](https:\/\/en.wikipedia.org\/wiki\/Category:Short_description_is_different_from_Wikidata \"Category:Short description is different from Wikidata\")\n- [Articles needing additional references from May 2024](https:\/\/en.wikipedia.org\/wiki\/Category:Articles_needing_additional_references_from_May_2024 \"Category:Articles needing additional references from May 2024\")\n- [All articles needing additional references](https:\/\/en.wikipedia.org\/wiki\/Category:All_articles_needing_additional_references \"Category:All articles needing additional references\")\n- [Wikipedia articles needing rewrite from July 2025](https:\/\/en.wikipedia.org\/wiki\/Category:Wikipedia_articles_needing_rewrite_from_July_2025 \"Category:Wikipedia articles needing rewrite from July 2025\")\n- [All articles needing rewrite](https:\/\/en.wikipedia.org\/wiki\/Category:All_articles_needing_rewrite \"Category:All articles needing rewrite\")\n- [Wikipedia articles needing reorganization from July 2025](https:\/\/en.wikipedia.org\/wiki\/Category:Wikipedia_articles_needing_reorganization_from_July_2025 \"Category:Wikipedia articles needing reorganization from July 2025\")\n- [Articles with multiple maintenance issues](https:\/\/en.wikipedia.org\/wiki\/Category:Articles_with_multiple_maintenance_issues \"Category:Articles with multiple maintenance issues\")\n- [Commons category link from Wikidata](https:\/\/en.wikipedia.org\/wiki\/Category:Commons_category_link_from_Wikidata \"Category:Commons category link from Wikidata\")\n\n- This page was last edited on 12 March 2026, at 18:37 (UTC).\n- Text is available under the [Creative Commons Attribution-ShareAlike 4.0 License](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Text_of_the_Creative_Commons_Attribution-ShareAlike_4.0_International_License \"Wikipedia:Text of the Creative Commons Attribution-ShareAlike 4.0 International License\"); additional terms may apply. By using this site, you agree to the [Terms of Use](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Policy:Terms_of_Use \"foundation:Special:MyLanguage\/Policy:Terms of Use\") and [Privacy Policy](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Policy:Privacy_policy \"foundation:Special:MyLanguage\/Policy:Privacy policy\"). Wikipedia® is a registered trademark of the [Wikimedia Foundation, Inc.](https:\/\/wikimediafoundation.org\/), a non-profit organization.\n\n- [Privacy policy](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Policy:Privacy_policy)\n- [About Wikipedia](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:About)\n- [Disclaimers](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:General_disclaimer)\n- [Contact Wikipedia](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Contact_us)\n- [Legal & safety contacts](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Legal:Wikimedia_Foundation_Legal_and_Safety_Contact_Information)\n- [Code of Conduct](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Policy:Universal_Code_of_Conduct)\n- [Developers](https:\/\/developer.wikimedia.org\/)\n- [Statistics](https:\/\/stats.wikimedia.org\/#\/en.wikipedia.org)\n- [Cookie statement](https:\/\/foundation.wikimedia.org\/wiki\/Special:MyLanguage\/Policy:Cookie_statement)\n- [Mobile view](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&mobileaction=toggle_view_mobile)\n\n- [![Wikimedia Foundation](https:\/\/en.wikipedia.org\/static\/images\/footer\/wikimedia.svg)](https:\/\/www.wikimedia.org\/)\n- [![Powered by MediaWiki](https:\/\/en.wikipedia.org\/w\/resources\/assets\/mediawiki_compact.svg)](https:\/\/www.mediawiki.org\/)\n\nSearch\n\nToggle the table of contents\n\nNatural language processing\n\n71 languages\n\n[Add topic](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing)","attrs_readable_markdown":"From Wikipedia, the free encyclopedia\n\n**Natural language processing** (**NLP**) is the processing of [natural language](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") information by a [computer](https:\/\/en.wikipedia.org\/wiki\/Computer \"Computer\"). NLP is a subfield of [computer science](https:\/\/en.wikipedia.org\/wiki\/Computer_science \"Computer science\") and is closely associated with [artificial intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence \"Artificial intelligence\"). NLP is also related to [information retrieval](https:\/\/en.wikipedia.org\/wiki\/Information_retrieval \"Information retrieval\"), [knowledge representation](https:\/\/en.wikipedia.org\/wiki\/Knowledge_representation_and_reasoning \"Knowledge representation and reasoning\"), [computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\"), and [linguistics](https:\/\/en.wikipedia.org\/wiki\/Linguistics \"Linguistics\") more broadly.[\\[1\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-nlpintro-1)\n\nMajor processing tasks in an NLP system include: [speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\"), [text classification](https:\/\/en.wikipedia.org\/wiki\/Text_classification \"Text classification\"), [natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_understanding \"Natural language understanding\"), and [natural language generation](https:\/\/en.wikipedia.org\/wiki\/Natural_language_generation \"Natural language generation\").\n\nNatural language processing has its roots in the 1950s.[\\[2\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-2) Already in 1950, [Alan Turing](https:\/\/en.wikipedia.org\/wiki\/Alan_Turing \"Alan Turing\") published an article titled \"[Computing Machinery and Intelligence](https:\/\/en.wikipedia.org\/wiki\/Computing_Machinery_and_Intelligence \"Computing Machinery and Intelligence\")\" which proposed what is now called the [Turing test](https:\/\/en.wikipedia.org\/wiki\/Turing_test \"Turing test\") as a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language.\n\n### Symbolic NLP (1950s – early 1990s)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=2 \"Edit section: Symbolic NLP (1950s – early 1990s)\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/0\/09\/AST_Document.svg\/250px-AST_Document.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:AST_Document.svg)\n\nA document parsed into an abstract syntax tree\n\nThe premise of symbolic NLP is often illustrated using [John Searle's Chinese room](https:\/\/en.wikipedia.org\/wiki\/John_Searle_\\(American_philosopher\\) \"John Searle (American philosopher)\") thought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts.\n\n- **1950s**: The [Georgetown experiment](https:\/\/en.wikipedia.org\/wiki\/Georgetown-IBM_experiment \"Georgetown-IBM experiment\") in 1954 involved fully [automatic translation](https:\/\/en.wikipedia.org\/wiki\/Automatic_translation \"Automatic translation\") of more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem.[\\[3\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-3) However, real progress was much slower, and after the [ALPAC report](https:\/\/en.wikipedia.org\/wiki\/ALPAC \"ALPAC\") in 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe[\\[4\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-4)) until the late 1980s when the first [statistical machine translation](https:\/\/en.wikipedia.org\/wiki\/Statistical_machine_translation \"Statistical machine translation\") systems were developed.\n- **1960s**: Some notably successful natural language processing systems developed in the 1960s were [SHRDLU](https:\/\/en.wikipedia.org\/wiki\/SHRDLU \"SHRDLU\"), a natural language system working in restricted \"[blocks worlds](https:\/\/en.wikipedia.org\/wiki\/Blocks_world \"Blocks world\")\" with restricted vocabularies, and [ELIZA](https:\/\/en.wikipedia.org\/wiki\/ELIZA \"ELIZA\"), a simulation of a [Rogerian psychotherapy](https:\/\/en.wikipedia.org\/wiki\/Rogerian_psychotherapy \"Rogerian psychotherapy\"), written by [Joseph Weizenbaum](https:\/\/en.wikipedia.org\/wiki\/Joseph_Weizenbaum \"Joseph Weizenbaum\") between 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the \"patient\" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to \"My head hurts\" with \"Why do you say your head hurts?\". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only *twenty* words, because that was all that would fit in a computer memory at the time.[\\[5\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-5)\n- **1970s**: During the 1970s, many programmers began to write \"conceptual [ontologies](https:\/\/en.wikipedia.org\/wiki\/Ontology_\\(information_science\\) \"Ontology (information science)\")\", which structured real-world information into computer-understandable data. Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981). During this time, the first [chatterbots](https:\/\/en.wikipedia.org\/wiki\/Chatterbots \"Chatterbots\") were written (e.g., [PARRY](https:\/\/en.wikipedia.org\/wiki\/PARRY \"PARRY\")).\n- **1980s**: The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of [HPSG](https:\/\/en.wikipedia.org\/wiki\/Head-driven_phrase_structure_grammar \"Head-driven phrase structure grammar\") as a computational operationalization of [generative grammar](https:\/\/en.wikipedia.org\/wiki\/Generative_grammar \"Generative grammar\")), morphology (e.g., two-level morphology[\\[6\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-6)), semantics (e.g., [Lesk algorithm](https:\/\/en.wikipedia.org\/wiki\/Lesk_algorithm \"Lesk algorithm\")), reference (e.g., within Centering Theory[\\[7\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-7)) and other areas of natural language understanding (e.g., in the [Rhetorical Structure Theory](https:\/\/en.wikipedia.org\/wiki\/Rhetorical_structure_theory \"Rhetorical structure theory\")). Other lines of research were continued, e.g., the development of chatterbots with [Racter](https:\/\/en.wikipedia.org\/wiki\/Racter \"Racter\") and [Jabberwacky](https:\/\/en.wikipedia.org\/wiki\/Jabberwacky \"Jabberwacky\"). An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period.[\\[8\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-8)\n\n### Statistical NLP (1990s–present)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=3 \"Edit section: Statistical NLP (1990s–present)\")\\]\n\nUp until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of [machine learning](https:\/\/en.wikipedia.org\/wiki\/Machine_learning \"Machine learning\") algorithms for language processing. This shift was influenced by increasing computational power (see [Moore's law](https:\/\/en.wikipedia.org\/wiki\/Moore%27s_law \"Moore's law\")) and a decline in the dominance of [Chomskyan](https:\/\/en.wikipedia.org\/wiki\/Chomskyan \"Chomskyan\") linguistic theories... (e.g. [transformational grammar](https:\/\/en.wikipedia.org\/wiki\/Transformational_grammar \"Transformational grammar\")), whose theoretical underpinnings discouraged the sort of [corpus linguistics](https:\/\/en.wikipedia.org\/wiki\/Corpus_linguistics \"Corpus linguistics\") that underlies the machine-learning approach to language processing.[\\[9\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-9)\n\n- **1990s**: Many of the notable early successes in statistical methods in NLP occurred in the field of [machine translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\"), due especially to work at IBM Research, such as [IBM alignment models](https:\/\/en.wikipedia.org\/wiki\/IBM_alignment_models \"IBM alignment models\"). These systems were able to take advantage of existing multilingual [textual corpora](https:\/\/en.wikipedia.org\/wiki\/Text_corpus \"Text corpus\") that had been produced by the [Parliament of Canada](https:\/\/en.wikipedia.org\/wiki\/Parliament_of_Canada \"Parliament of Canada\") and the [European Union](https:\/\/en.wikipedia.org\/wiki\/European_Union \"European Union\") as a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data.\n- **2000s**: With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on [unsupervised](https:\/\/en.wikipedia.org\/wiki\/Unsupervised_learning \"Unsupervised learning\") and [semi-supervised learning](https:\/\/en.wikipedia.org\/wiki\/Semi-supervised_learning \"Semi-supervised learning\") algorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than [supervised learning](https:\/\/en.wikipedia.org\/wiki\/Supervised_learning \"Supervised learning\"), and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the [World Wide Web](https:\/\/en.wikipedia.org\/wiki\/World_Wide_Web \"World Wide Web\")), which can often make up for the worse efficiency if the algorithm used has a low enough [time complexity](https:\/\/en.wikipedia.org\/wiki\/Time_complexity \"Time complexity\") to be practical.\n- **2003:** [word n-gram model](https:\/\/en.wikipedia.org\/wiki\/Word_n-gram_language_model \"Word n-gram language model\"), at the time the best statistical algorithm, is outperformed by a [multi-layer perceptron](https:\/\/en.wikipedia.org\/wiki\/Multi-layer_perceptron \"Multi-layer perceptron\") (with a single hidden layer and context length of several words, trained on up to 14 million words, by [Bengio](https:\/\/en.wikipedia.org\/wiki\/Yoshua_Bengio \"Yoshua Bengio\") et al.)[\\[10\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-10)\n- **2010:** [Tomáš Mikolov](https:\/\/en.wikipedia.org\/wiki\/Tom%C3%A1%C5%A1_Mikolov \"Tomáš Mikolov\") (then a PhD student at [Brno University of Technology](https:\/\/en.wikipedia.org\/wiki\/Brno_University_of_Technology \"Brno University of Technology\")) with co-authors applied a simple [recurrent neural network](https:\/\/en.wikipedia.org\/wiki\/Recurrent_neural_network \"Recurrent neural network\") with a single hidden layer to language modeling,[\\[11\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-11) and in the following years he went on to develop [Word2vec](https:\/\/en.wikipedia.org\/wiki\/Word2vec \"Word2vec\"). In the 2010s, [representation learning](https:\/\/en.wikipedia.org\/wiki\/Representation_learning \"Representation learning\") and [deep neural network](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\")\\-style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques[\\[12\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-goldberg:nnlp17-12)[\\[13\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-goodfellow:book16-13) can achieve state-of-the-art results in many natural language tasks, e.g., in [language modeling](https:\/\/en.wikipedia.org\/wiki\/Language_modeling \"Language modeling\")[\\[14\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-jozefowicz:lm16-14) and parsing.[\\[15\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-choe:emnlp16-15)[\\[16\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-vinyals:nips15-16) This is increasingly important [in medicine and healthcare](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_in_healthcare \"Artificial intelligence in healthcare\"), where NLP helps analyze notes and text in [electronic health records](https:\/\/en.wikipedia.org\/wiki\/Electronic_health_record \"Electronic health record\") that would otherwise be inaccessible for study when seeking to improve care[\\[17\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-17) or protect patient privacy.[\\[18\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-18)\n\n## Approaches: Symbolic, statistical, neural networks\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=4 \"Edit section: Approaches: Symbolic, statistical, neural networks\")\\]\n\nSymbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular:[\\[19\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-winograd:shrdlu71-19)[\\[20\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-schank77-20) such as by writing grammars or devising heuristic rules for [stemming](https:\/\/en.wikipedia.org\/wiki\/Stemming \"Stemming\").\n\n[Machine learning](https:\/\/en.wikipedia.org\/wiki\/Machine_learning \"Machine learning\") approaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach:\n\n- both statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally.\n- [language models](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\"), produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce.\n- the larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to [intractability](https:\/\/en.wikipedia.org\/wiki\/Intractable_problem \"Intractable problem\") problems.\n\nRule-based systems are commonly used:\n\n- when the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the [Apertium](https:\/\/en.wikipedia.org\/wiki\/Apertium \"Apertium\") system,\n- for preprocessing in NLP pipelines, e.g., [tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(lexical_analysis\\) \"Tokenization (lexical analysis)\"), or\n- for post-processing and transforming the output of NLP pipelines, e.g., for [knowledge extraction](https:\/\/en.wikipedia.org\/wiki\/Knowledge_extraction \"Knowledge extraction\") from syntactic parses.\n\n### Statistical approach\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=5 \"Edit section: Statistical approach\")\\]\n\nIn the late 1980s and mid-1990s, the statistical approach ended a period of [AI winter](https:\/\/en.wikipedia.org\/wiki\/AI_winter \"AI winter\"), which was caused by the inefficiencies of the rule-based approaches.[\\[21\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-johnson:eacl:ilcl09-21)[\\[22\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-resnik:langlog11-22)\n\nThe earliest [decision trees](https:\/\/en.wikipedia.org\/wiki\/Decision_tree \"Decision tree\"), producing systems of hard [if–then rules](https:\/\/en.wikipedia.org\/wiki\/Conditional_\\(computer_programming\\)#If%E2%80%93then\\(%E2%80%93else\\) \"Conditional (computer programming)\"), were still very similar to the old rule-based approaches. Only the introduction of hidden [Markov models](https:\/\/en.wikipedia.org\/wiki\/Markov_model \"Markov model\"), applied to part-of-speech tagging, announced the end of the old rule-based approach.\n\nA major drawback of statistical methods is that they require elaborate [feature engineering](https:\/\/en.wikipedia.org\/wiki\/Feature_engineering \"Feature engineering\"). Since 2015,[\\[23\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-23) [neural network](https:\/\/en.wikipedia.org\/wiki\/Neural_network_\\(machine_learning\\) \"Neural network (machine learning)\")–based methods have increasingly replaced traditional statistical approaches, using [semantic networks](https:\/\/en.wikipedia.org\/wiki\/Semantic_networks \"Semantic networks\")[\\[24\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-24) and [word embeddings](https:\/\/en.wikipedia.org\/wiki\/Word_embedding \"Word embedding\") to capture semantic properties of words.\n\nIntermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore.\n\n[Neural machine translation](https:\/\/en.wikipedia.org\/wiki\/Neural_machine_translation \"Neural machine translation\"), based on then-newly invented [sequence-to-sequence](https:\/\/en.wikipedia.org\/wiki\/Seq2seq \"Seq2seq\") transformations, made obsolete the intermediate steps, such as word alignment, previously necessary for [statistical machine translation](https:\/\/en.wikipedia.org\/wiki\/Statistical_machine_translation \"Statistical machine translation\").\n\nThe following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks.\n\nThough natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below.\n\n### Text and speech processing\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=8 \"Edit section: Text and speech processing\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/8\/8c\/Hebrew_stop_words_word_cloud.png\/250px-Hebrew_stop_words_word_cloud.png)](https:\/\/en.wikipedia.org\/wiki\/File:Hebrew_stop_words_word_cloud.png)\n\nWord cloud of stop words in Hebrew\n\n[Optical character recognition](https:\/\/en.wikipedia.org\/wiki\/Optical_character_recognition \"Optical character recognition\") (OCR)\n\nGiven an image representing printed text, determine the corresponding text.\n\n[Speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\")\n\nGiven a sound clip of a person or people speaking, determine the textual representation of the speech. This is the opposite of [text to speech](https:\/\/en.wikipedia.org\/wiki\/Text_to_speech \"Text to speech\") and is one of the extremely difficult problems colloquially termed \"[AI-complete](https:\/\/en.wikipedia.org\/wiki\/AI-complete \"AI-complete\")\" (see above). In [natural speech](https:\/\/en.wikipedia.org\/wiki\/Natural_speech \"Natural speech\") there are hardly any pauses between successive words, and thus [speech segmentation](https:\/\/en.wikipedia.org\/wiki\/Speech_segmentation \"Speech segmentation\") is a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed [coarticulation](https:\/\/en.wikipedia.org\/wiki\/Coarticulation \"Coarticulation\"), so the conversion of the [analog signal](https:\/\/en.wikipedia.org\/wiki\/Analog_signal \"Analog signal\") to discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent.\n\n[Speech segmentation](https:\/\/en.wikipedia.org\/wiki\/Speech_segmentation \"Speech segmentation\")\n\nGiven a sound clip of a person or people speaking, separate it into words. A subtask of [speech recognition](https:\/\/en.wikipedia.org\/wiki\/Speech_recognition \"Speech recognition\") and typically grouped with it.\n\n[Text-to-speech](https:\/\/en.wikipedia.org\/wiki\/Text-to-speech \"Text-to-speech\")\n\nGiven a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired.[\\[25\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-25)\n\n[Word segmentation](https:\/\/en.wikipedia.org\/wiki\/Word_segmentation \"Word segmentation\") ([Tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(lexical_analysis\\) \"Tokenization (lexical analysis)\"))\n\n[Tokenization](https:\/\/en.wikipedia.org\/wiki\/Tokenization_\\(data_security\\) \"Tokenization (data security)\") is a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods.[\\[26\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-:0-26)\n\nFor a language like [English](https:\/\/en.wikipedia.org\/wiki\/English_language \"English language\"), this is fairly trivial, since words are usually separated by spaces. However, some written languages like [Chinese](https:\/\/en.wikipedia.org\/wiki\/Chinese_language \"Chinese language\"), [Japanese](https:\/\/en.wikipedia.org\/wiki\/Japanese_language \"Japanese language\") and [Thai](https:\/\/en.wikipedia.org\/wiki\/Thai_language \"Thai language\") do not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the [vocabulary](https:\/\/en.wikipedia.org\/wiki\/Vocabulary \"Vocabulary\") and [morphology](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\") of words in the language. Sometimes this process is also used in cases like [bag of words](https:\/\/en.wikipedia.org\/wiki\/Bag_of_words \"Bag of words\") (BOW) creation in data mining.[\\[27\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-27)[\\[28\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-28)\n\n### Morphological analysis\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=9 \"Edit section: Morphological analysis\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/2\/27\/Eustagger_euskal_lematizatzailearen_adibidea.png\/250px-Eustagger_euskal_lematizatzailearen_adibidea.png)](https:\/\/en.wikipedia.org\/wiki\/File:Eustagger_euskal_lematizatzailearen_adibidea.png)\n\nLemmatization of Basque words\n\n[Lemmatization](https:\/\/en.wikipedia.org\/wiki\/Lemmatisation \"Lemmatisation\")\n\nThe task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form.[\\[29\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-29)\n\n[Morphological segmentation](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\")\n\nSeparate words into individual [morphemes](https:\/\/en.wikipedia.org\/wiki\/Morpheme \"Morpheme\") and identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the [morphology](https:\/\/en.wikipedia.org\/wiki\/Morphology_\\(linguistics\\) \"Morphology (linguistics)\") (*i.e.*, the structure of words) of the language being considered. [English](https:\/\/en.wikipedia.org\/wiki\/English_language \"English language\") has fairly simple morphology, especially [inflectional morphology](https:\/\/en.wikipedia.org\/wiki\/Inflectional_morphology \"Inflectional morphology\"), and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., \"open, opens, opened, opening\") as separate words. In languages such as [Turkish](https:\/\/en.wikipedia.org\/wiki\/Turkish_language \"Turkish language\") or [Meitei](https:\/\/en.wikipedia.org\/wiki\/Meitei_language \"Meitei language\"), a highly [agglutinated](https:\/\/en.wikipedia.org\/wiki\/Agglutination \"Agglutination\") Indian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms.[\\[30\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-30)\n\n[Part-of-speech tagging](https:\/\/en.wikipedia.org\/wiki\/Part-of-speech_tagging \"Part-of-speech tagging\")\n\nGiven a sentence, determine the [part of speech](https:\/\/en.wikipedia.org\/wiki\/Part_of_speech \"Part of speech\") (POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, \"book\" can be a [noun](https:\/\/en.wikipedia.org\/wiki\/Noun \"Noun\") (\"the book on the table\") or [verb](https:\/\/en.wikipedia.org\/wiki\/Verb \"Verb\") (\"to book a flight\"); \"set\" can be a noun, verb or [adjective](https:\/\/en.wikipedia.org\/wiki\/Adjective \"Adjective\"); and \"out\" can be any of at least five different parts of speech.\n\n[Stemming](https:\/\/en.wikipedia.org\/wiki\/Stemming \"Stemming\")\n\nThe process of reducing inflected (or sometimes derived) words to a base form (e.g., \"close\" will be the root for \"closed\", \"closing\", \"close\", \"closer\" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary.\n\n[Grammar induction](https:\/\/en.wikipedia.org\/wiki\/Grammar_induction \"Grammar induction\")[\\[31\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-31)\n\nGenerate a [formal grammar](https:\/\/en.wikipedia.org\/wiki\/Formal_grammar \"Formal grammar\") that describes a language's syntax.\n\n[Sentence breaking](https:\/\/en.wikipedia.org\/wiki\/Sentence_breaking \"Sentence breaking\") (also known as \"[sentence boundary disambiguation](https:\/\/en.wikipedia.org\/wiki\/Sentence_boundary_disambiguation \"Sentence boundary disambiguation\")\")\n\nGiven a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by [periods](https:\/\/en.wikipedia.org\/wiki\/Full_stop \"Full stop\") or other [punctuation marks](https:\/\/en.wikipedia.org\/wiki\/Punctuation_mark \"Punctuation mark\"), but these same characters can serve other purposes (e.g., marking [abbreviations](https:\/\/en.wikipedia.org\/wiki\/Abbreviation \"Abbreviation\")).\n\n[Parsing](https:\/\/en.wikipedia.org\/wiki\/Parsing \"Parsing\")\n\nDetermine the [parse tree](https:\/\/en.wikipedia.org\/wiki\/Parse_tree \"Parse tree\") (grammatical analysis) of a given sentence. The [grammar](https:\/\/en.wikipedia.org\/wiki\/Grammar \"Grammar\") for [natural languages](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") is [ambiguous](https:\/\/en.wikipedia.org\/wiki\/Ambiguous \"Ambiguous\") and typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing: *dependency parsing* and *constituency parsing*. Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a [probabilistic context-free grammar](https:\/\/en.wikipedia.org\/wiki\/Probabilistic_context-free_grammar \"Probabilistic context-free grammar\") (PCFG) (see also *[stochastic grammar](https:\/\/en.wikipedia.org\/wiki\/Stochastic_grammar \"Stochastic grammar\")*).\n\n### Lexical semantics (of individual words in context)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=11 \"Edit section: Lexical semantics (of individual words in context)\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/3\/34\/Entity_Linking_-_Example_of_pipeline.png\/250px-Entity_Linking_-_Example_of_pipeline.png)](https:\/\/en.wikipedia.org\/wiki\/File:Entity_Linking_-_Example_of_pipeline.png)\n\nAn entity linking pipeline\n\n[Lexical semantics](https:\/\/en.wikipedia.org\/wiki\/Lexical_semantics \"Lexical semantics\")\n\nWhat is the computational meaning of individual words in context?\n\n[Distributional semantics](https:\/\/en.wikipedia.org\/wiki\/Distributional_semantics \"Distributional semantics\")\n\nHow can we learn semantic representations from data?\n\n[Named entity recognition](https:\/\/en.wikipedia.org\/wiki\/Named_entity_recognition \"Named entity recognition\") (NER)\n\nGiven a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although [capitalization](https:\/\/en.wikipedia.org\/wiki\/Capitalization \"Capitalization\") can aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of [named entity](https:\/\/en.wikipedia.org\/wiki\/Named_entity \"Named entity\"), and in any case, is often inaccurate or insufficient. For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized. Furthermore, many other languages in non-Western scripts (e.g. [Chinese](https:\/\/en.wikipedia.org\/wiki\/Chinese_language \"Chinese language\") or [Arabic](https:\/\/en.wikipedia.org\/wiki\/Arabic_language \"Arabic language\")) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example, [German](https:\/\/en.wikipedia.org\/wiki\/German_language \"German language\") capitalizes all [nouns](https:\/\/en.wikipedia.org\/wiki\/Noun \"Noun\"), regardless of whether they are names, and [French](https:\/\/en.wikipedia.org\/wiki\/French_language \"French language\") and [Spanish](https:\/\/en.wikipedia.org\/wiki\/Spanish_language \"Spanish language\") do not capitalize names that serve as [adjectives](https:\/\/en.wikipedia.org\/wiki\/Adjective \"Adjective\"). This task is also referred to as token classification.[\\[32\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-32)\n\n[Sentiment analysis](https:\/\/en.wikipedia.org\/wiki\/Sentiment_analysis \"Sentiment analysis\") (see also [Multimodal sentiment analysis](https:\/\/en.wikipedia.org\/wiki\/Multimodal_sentiment_analysis \"Multimodal sentiment analysis\"))\n\nSentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as [word n-grams](https:\/\/en.wikipedia.org\/wiki\/Word_n-gram_language_model \"Word n-gram language model\"), [Term Frequency-Inverse Document Frequency](https:\/\/en.wikipedia.org\/wiki\/Term_frequency-inverse_document_frequency \"Term frequency-inverse document frequency\") (TF-IDF) features, hand-generated features, or employ [deep learning](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\") models designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms.[\\[26\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-:0-26)\n\n[Terminology extraction](https:\/\/en.wikipedia.org\/wiki\/Terminology_extraction \"Terminology extraction\")\n\nThe goal of terminology extraction is to automatically extract relevant terms from a given corpus.\n\n[Word-sense disambiguation](https:\/\/en.wikipedia.org\/wiki\/Word-sense_disambiguation \"Word-sense disambiguation\") (WSD)\n\nMany words have more than one [meaning](https:\/\/en.wikipedia.org\/wiki\/Meaning_\\(linguistics\\) \"Meaning (linguistics)\"); we have to select the meaning which makes the most sense in context. For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as [WordNet](https:\/\/en.wikipedia.org\/wiki\/WordNet \"WordNet\").\n\n[Entity linking](https:\/\/en.wikipedia.org\/wiki\/Entity_linking \"Entity linking\")\n\nMany words—typically proper names—refer to [named entities](https:\/\/en.wikipedia.org\/wiki\/Named_entity \"Named entity\"); here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context.\n\n### Relational semantics (semantics of individual sentences)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=12 \"Edit section: Relational semantics (semantics of individual sentences)\")\\]\n\n[Relationship extraction](https:\/\/en.wikipedia.org\/wiki\/Relationship_extraction \"Relationship extraction\")\n\nGiven a chunk of text, identify the relationships among named entities (e.g. who is married to whom).\n\n[Semantic parsing](https:\/\/en.wikipedia.org\/wiki\/Semantic_parsing \"Semantic parsing\")\n\nGiven a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in [AMR parsing](https:\/\/en.wikipedia.org\/wiki\/Abstract_Meaning_Representation \"Abstract Meaning Representation\")) or in accordance with a logical formalism (e.g., in [DRT parsing](https:\/\/en.wikipedia.org\/wiki\/Discourse_representation_theory \"Discourse representation theory\")). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see [Natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Natural_language_understanding) below).\n\n[Semantic role labelling](https:\/\/en.wikipedia.org\/wiki\/Semantic_role_labeling \"Semantic role labeling\") (see also implicit semantic role labelling below)\n\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https:\/\/en.wikipedia.org\/wiki\/Frame_semantics_\\(linguistics\\) \"Frame semantics (linguistics)\")), then identify and classify the frame elements ([semantic roles](https:\/\/en.wikipedia.org\/wiki\/Semantic_roles \"Semantic roles\")).\n\n### Discourse (semantics beyond individual sentences)\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=13 \"Edit section: Discourse (semantics beyond individual sentences)\")\\]\n\n[Coreference resolution](https:\/\/en.wikipedia.org\/wiki\/Coreference \"Coreference\")\n\nGiven a sentence or larger chunk of text, determine which words (\"mentions\") refer to the same objects (\"entities\"). [Anaphora resolution](https:\/\/en.wikipedia.org\/wiki\/Anaphora_resolution \"Anaphora resolution\") is a specific example of this task, and is specifically concerned with matching up [pronouns](https:\/\/en.wikipedia.org\/wiki\/Pronoun \"Pronoun\") with the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called \"bridging relationships\" involving [referring expressions](https:\/\/en.wikipedia.org\/wiki\/Referring_expression \"Referring expression\"). For example, in a sentence such as \"He entered John's house through the front door\", \"the front door\" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to).\n\n[Discourse analysis](https:\/\/en.wikipedia.org\/wiki\/Discourse_analysis \"Discourse analysis\")\n\nThis rubric includes several related tasks. One task is discourse parsing, i.e., identifying the [discourse](https:\/\/en.wikipedia.org\/wiki\/Discourse \"Discourse\") structure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast). Another possible task is recognizing and classifying the [speech acts](https:\/\/en.wikipedia.org\/wiki\/Speech_act \"Speech act\") in a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.).\n\nImplicit semantic role labelling\n\nGiven a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https:\/\/en.wikipedia.org\/wiki\/Frame_semantics_\\(linguistics\\) \"Frame semantics (linguistics)\")) and their explicit semantic roles in the current sentence (see [Semantic role labelling](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Semantic_role_labelling) above). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to [pro-drop languages](https:\/\/en.wikipedia.org\/wiki\/Pro-drop_language \"Pro-drop language\").\n\n[Recognizing textual entailment](https:\/\/en.wikipedia.org\/wiki\/Textual_entailment \"Textual entailment\")\n\nGiven two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false.[\\[33\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-rte:11-33)\n\n[Topic segmentation](https:\/\/en.wikipedia.org\/wiki\/Topic_segmentation \"Topic segmentation\") and recognition\n\nGiven a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment.\n\n[Argument mining](https:\/\/en.wikipedia.org\/wiki\/Argument_mining \"Argument mining\")\n\nThe goal of argument mining is the automatic extraction and identification of argumentative structures from [natural language](https:\/\/en.wikipedia.org\/wiki\/Natural_language \"Natural language\") text with the aid of computer programs.[\\[34\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-34) Such argumentative structures include the premise, conclusions, the [argument scheme](https:\/\/en.wikipedia.org\/wiki\/Argument_scheme \"Argument scheme\") and the relationship between the main and subsidiary argument, or the main and counter-argument within discourse.[\\[35\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-35)[\\[36\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-36)\n\n### Higher-level NLP applications\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=14 \"Edit section: Higher-level NLP applications\")\\]\n\n[![](https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/4\/43\/Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png\/250px-Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)](https:\/\/en.wikipedia.org\/wiki\/File:Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)\n\nMachine translation in Firefox\n\n[Automatic summarization](https:\/\/en.wikipedia.org\/wiki\/Automatic_summarization \"Automatic summarization\") (text summarization)\n\nProduce a readable summary of a chunk of text. Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper.\n\nGrammatical error correction\n\nGrammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology\/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011.[\\[37\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-37)[\\[38\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-38)[\\[39\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-39) As far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as [GPT-2](https:\/\/en.wikipedia.org\/wiki\/GPT-2 \"GPT-2\"), this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications.\n\n[Logic translation](https:\/\/en.wikipedia.org\/wiki\/Logic_translation \"Logic translation\")\n\nTranslate a text from a natural language into formal logic.\n\n[Machine translation](https:\/\/en.wikipedia.org\/wiki\/Machine_translation \"Machine translation\") (MT)\n\nAutomatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed \"[AI-complete](https:\/\/en.wikipedia.org\/wiki\/AI-complete \"AI-complete\")\", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly.\n\n[Natural language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural_language_understanding \"Natural language understanding\") (NLU)\n\nConvert chunks of text into more formal representations such as [first-order logic](https:\/\/en.wikipedia.org\/wiki\/First-order_logic \"First-order logic\") structures that are easier for [computer](https:\/\/en.wikipedia.org\/wiki\/Computer \"Computer\") programs to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as [closed-world assumption](https:\/\/en.wikipedia.org\/wiki\/Closed-world_assumption \"Closed-world assumption\") (CWA) vs. [open-world assumption](https:\/\/en.wikipedia.org\/wiki\/Open-world_assumption \"Open-world assumption\"), or subjective Yes\/No vs. objective True\/False is expected for the construction of a basis of semantics formalization.[\\[40\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-40)\n\n[Natural language generation](https:\/\/en.wikipedia.org\/wiki\/Natural_language_generation \"Natural language generation\") (NLG):\n\nConvert information from computer databases or semantic intents into readable human language.\n\nBook generation\n\nNot an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter, *The policeman's beard is half-constructed*).[\\[41\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-41) The first published work by a neural network was published in 2018, *[1 the Road](https:\/\/en.wikipedia.org\/wiki\/1_the_Road \"1 the Road\")*, marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free) [language models](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\"). The first machine-generated science book was published in 2019 (Beta Writer, *Lithium-Ion Batteries*, Springer, Cham).[\\[42\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-42) Unlike *Racter* and *1 the Road*, this is grounded on factual knowledge and based on text summarization.\n\n[Document AI](https:\/\/en.wikipedia.org\/wiki\/Document_AI \"Document AI\")\n\nA Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants.[\\[43\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-43)\n\n[Dialogue management](https:\/\/en.wikipedia.org\/wiki\/Dialogue_system \"Dialogue system\")\n\nComputer systems intended to converse with a human.\n\n[Question answering](https:\/\/en.wikipedia.org\/wiki\/Question_answering \"Question answering\")\n\nGiven a human-language question, determine its answer. Typical questions have a specific right answer (such as \"What is the capital of Canada?\"), but sometimes open-ended questions are also considered (such as \"What is the meaning of life?\").\n\n[Text-to-image generation](https:\/\/en.wikipedia.org\/wiki\/Text-to-image_generation \"Text-to-image generation\")\n\nGiven a description of an image, generate an image that matches the description.[\\[44\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-44)\n\nText-to-scene generation\n\nGiven a description of a scene, generate a [3D model](https:\/\/en.wikipedia.org\/wiki\/3D_model \"3D model\") of the scene.[\\[45\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-45)[\\[46\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-46)\n\n[Text-to-video](https:\/\/en.wikipedia.org\/wiki\/Text-to-video_model \"Text-to-video model\")\n\nGiven a description of a video, generate a video that matches the description.[\\[47\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-47)[\\[48\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-48)\n\n## General tendencies and (possible) future directions\n\\[[edit](https:\/\/en.wikipedia.org\/w\/index.php?title=Natural_language_processing&action=edit&section=15 \"Edit section: General tendencies and (possible) future directions\")\\]\n\nBased on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed:[\\[49\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-49)\n\n- Interest on increasingly abstract, \"cognitive\" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09\/2017–18: dependency syntax, 2004–05\/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing).\n- Increasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+\/100+ languages)\n- Elimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems)\n\nMost higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above).\n\n[Cognition](https:\/\/en.wikipedia.org\/wiki\/Cognition \"Cognition\") refers to \"the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses.\"[\\[50\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-50) [Cognitive science](https:\/\/en.wikipedia.org\/wiki\/Cognitive_science \"Cognitive science\") is the interdisciplinary, scientific study of the mind and its processes.[\\[51\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-51) [Cognitive linguistics](https:\/\/en.wikipedia.org\/wiki\/Cognitive_linguistics \"Cognitive linguistics\") is an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics.[\\[52\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-52) Especially during the age of [symbolic NLP](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#Symbolic_NLP_\\(1950s_%E2%80%93_early_1990s\\)), the area of computational linguistics maintained strong ties with cognitive studies.\n\nAs an example, [George Lakoff](https:\/\/en.wikipedia.org\/wiki\/George_Lakoff \"George Lakoff\") offers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics,[\\[53\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-53) with two defining aspects:\n\n1. Apply the theory of [conceptual metaphor](https:\/\/en.wikipedia.org\/wiki\/Conceptual_metaphor \"Conceptual metaphor\"), explained by Lakoff as \"the understanding of one idea, in terms of another\" which provides an idea of the intent of the author.[\\[54\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-54) For example, consider the English word *big*. When used in a comparison (\"That is a big tree\"), the author's intent is to imply that the tree is *physically large* relative to other trees or the authors experience. When used metaphorically (\"Tomorrow is a big day\"), the author's intent to imply *importance*. The intent behind other usages, like in \"She is a big person\", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information.\n2. Assign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a [probabilistic context-free grammar](https:\/\/en.wikipedia.org\/wiki\/Probabilistic_context-free_grammar \"Probabilistic context-free grammar\") (PCFG). The mathematical equation for such algorithms is presented in [US Patent 9269353](https:\/\/worldwide.espacenet.com\/patent\/search\/family\/055314712\/publication\/US9269353B1?q=pn%3DUS9269353):[\\[55\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-55)\n\n![{\\\\displaystyle {RMM(token\\_{N})}={PMM(token\\_{N})}\\\\times {\\\\frac {1}{2d}}\\\\left(\\\\sum \\_{i=-d}^{d}{((PMM(token\\_{N})}\\\\times {PF(token\\_{N-i},token\\_{N},token\\_{N+i}))\\_{i}}\\\\right)}](https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/43ccadd794a4b84e20d1209997a463342e0dfbfe)\n\n*Where*\n\n**RMM** is the relative measure of meaning\n\n**token** is any block of text, sentence, phrase or word\n\n**N** is the number of tokens being analyzed\n\n**PMM** is the probable measure of meaning based on a corpora\n\n**d** is the non zero location of the token along the sequence of **N** tokens\n\n**PF** is the probability function specific to a language\n\nTies with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar,[\\[56\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-56) functional grammar,[\\[57\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-57) construction grammar,[\\[58\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-58) computational psycholinguistics and cognitive neuroscience (e.g., [ACT-R](https:\/\/en.wikipedia.org\/wiki\/ACT-R \"ACT-R\")), however, with limited uptake in mainstream NLP (as measured by presence on major conferences[\\[59\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-59) of the [ACL](https:\/\/en.wikipedia.org\/wiki\/Association_for_Computational_Linguistics \"Association for Computational Linguistics\")). More recently, ideas of cognitive NLP have been revived as an approach to achieve [explainability](https:\/\/en.wikipedia.org\/wiki\/Explainable_artificial_intelligence \"Explainable artificial intelligence\"), e.g., under the notion of \"cognitive AI\".[\\[60\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-60) Likewise, ideas of cognitive NLP are inherent to neural models [multimodal](https:\/\/en.wikipedia.org\/wiki\/Multimodal_interaction \"Multimodal interaction\") NLP (although rarely made explicit)[\\[61\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-61) and developments in [artificial intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence \"Artificial intelligence\"), specifically tools and technologies using [large language model](https:\/\/en.wikipedia.org\/wiki\/Large_language_model \"Large language model\") approaches[\\[62\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-62) and new directions in [artificial general intelligence](https:\/\/en.wikipedia.org\/wiki\/Artificial_general_intelligence \"Artificial general intelligence\") based on the [free energy principle](https:\/\/en.wikipedia.org\/wiki\/Free_energy_principle \"Free energy principle\")[\\[63\\]](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_note-63) by British neuroscientist and theoretician at University College London [Karl J. Friston](https:\/\/en.wikipedia.org\/wiki\/Karl_J._Friston \"Karl J. Friston\").\n\n- *[1 the Road](https:\/\/en.wikipedia.org\/wiki\/1_the_Road \"1 the Road\")*\n- [Artificial intelligence detection software](https:\/\/en.wikipedia.org\/wiki\/Artificial_intelligence_detection_software \"Artificial intelligence detection software\")\n- [Automated essay scoring](https:\/\/en.wikipedia.org\/wiki\/Automated_essay_scoring \"Automated essay scoring\")\n- [Biomedical text mining](https:\/\/en.wikipedia.org\/wiki\/Biomedical_text_mining \"Biomedical text mining\")\n- [Compound term processing](https:\/\/en.wikipedia.org\/wiki\/Compound_term_processing \"Compound term processing\")\n- [Computational linguistics](https:\/\/en.wikipedia.org\/wiki\/Computational_linguistics \"Computational linguistics\")\n- [Computer-assisted reviewing](https:\/\/en.wikipedia.org\/wiki\/Computer-assisted_reviewing \"Computer-assisted reviewing\")\n- [Controlled natural language](https:\/\/en.wikipedia.org\/wiki\/Controlled_natural_language \"Controlled natural language\")\n- [Deep learning](https:\/\/en.wikipedia.org\/wiki\/Deep_learning \"Deep learning\")\n- [Deep linguistic processing](https:\/\/en.wikipedia.org\/wiki\/Deep_linguistic_processing \"Deep linguistic processing\")\n- [Distributional semantics](https:\/\/en.wikipedia.org\/wiki\/Distributional_semantics \"Distributional semantics\")\n- [Foreign language reading aid](https:\/\/en.wikipedia.org\/wiki\/Foreign_language_reading_aid \"Foreign language reading aid\")\n- [Foreign language writing aid](https:\/\/en.wikipedia.org\/wiki\/Foreign_language_writing_aid \"Foreign language writing aid\")\n- [Information extraction](https:\/\/en.wikipedia.org\/wiki\/Information_extraction \"Information extraction\")\n- [Information retrieval](https:\/\/en.wikipedia.org\/wiki\/Information_retrieval \"Information retrieval\")\n- [Language and Communication Technologies](https:\/\/en.wikipedia.org\/wiki\/Language_and_Communication_Technologies \"Language and Communication Technologies\")\n- [Language model](https:\/\/en.wikipedia.org\/wiki\/Language_model \"Language model\")\n- [Language technology](https:\/\/en.wikipedia.org\/wiki\/Language_technology \"Language technology\")\n- [Latent semantic indexing](https:\/\/en.wikipedia.org\/wiki\/Latent_semantic_indexing \"Latent semantic indexing\")\n- [Multi-agent system](https:\/\/en.wikipedia.org\/wiki\/Multi-agent_system \"Multi-agent system\")\n- [Native-language identification](https:\/\/en.wikipedia.org\/wiki\/Native-language_identification \"Native-language identification\")\n- [Natural-language programming](https:\/\/en.wikipedia.org\/wiki\/Natural-language_programming \"Natural-language programming\")\n- [Natural-language understanding](https:\/\/en.wikipedia.org\/wiki\/Natural-language_understanding \"Natural-language understanding\")\n- [Natural-language search](https:\/\/en.wikipedia.org\/wiki\/Natural-language_user_interface \"Natural-language user interface\")\n- [Outline of natural language processing](https:\/\/en.wikipedia.org\/wiki\/Outline_of_natural_language_processing \"Outline of natural language processing\")\n- [Query expansion](https:\/\/en.wikipedia.org\/wiki\/Query_expansion \"Query expansion\")\n- [Query understanding](https:\/\/en.wikipedia.org\/wiki\/Query_understanding \"Query understanding\")\n- [Reification (linguistics)](https:\/\/en.wikipedia.org\/wiki\/Reification_\\(linguistics\\) \"Reification (linguistics)\")\n- [Speech processing](https:\/\/en.wikipedia.org\/wiki\/Speech_processing \"Speech processing\")\n- [Spoken dialogue systems](https:\/\/en.wikipedia.org\/wiki\/Spoken_dialogue_systems \"Spoken dialogue systems\")\n- [Text-proofing](https:\/\/en.wikipedia.org\/wiki\/Text-proofing \"Text-proofing\")\n- [Text simplification](https:\/\/en.wikipedia.org\/wiki\/Text_simplification \"Text simplification\")\n- [Transformer (machine learning model)](https:\/\/en.wikipedia.org\/wiki\/Transformer_\\(machine_learning_model\\) \"Transformer (machine learning model)\")\n- [Truecasing](https:\/\/en.wikipedia.org\/wiki\/Truecasing \"Truecasing\")\n- [Question answering](https:\/\/en.wikipedia.org\/wiki\/Question_answering \"Question answering\")\n- [Word2vec](https:\/\/en.wikipedia.org\/wiki\/Word2vec \"Word2vec\")\n\n1. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-nlpintro_1-0)**\n   Eisenstein, Jacob (October 1, 2019). [*Introduction to Natural Language Processing*](https:\/\/mitpress.mit.edu\/9780262042840\/introduction-to-natural-language-processing\/). The MIT Press. p. 1. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n   \n   [978-0-262-04284-0](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-04284-0 \"Special:BookSources\/978-0-262-04284-0\")\n   \n   .\n2. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-2)**\n   [\"NLP\"](https:\/\/cs.stanford.edu\/people\/eroberts\/courses\/soco\/projects\/2004-05\/nlp\/overview_history.html).\n3. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-3)**\n   Hutchins, J. (2005). [\"The history of machine translation in a nutshell\"](https:\/\/web.archive.org\/web\/20190713103044\/http:\/\/www.hutchinsweb.me.uk\/Nutshell-2005.pdf) (PDF). Archived from [the original](http:\/\/www.hutchinsweb.me.uk\/Nutshell-2005.pdf) (PDF) on 2019-07-13. Retrieved 2019-02-04.\n   \n   \\[*[self-published source](https:\/\/en.wikipedia.org\/wiki\/Wikipedia:Verifiability#Self-published_sources \"Wikipedia:Verifiability\")*\\]\n4. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-4)** \"ALPAC: the (in)famous report\", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12.\n5. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-5)** [Crevier 1993](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#CITEREFCrevier1993), pp. 146–148, see also [Buchanan 2005](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#CITEREFBuchanan2005), p. 56: \"Early programs were necessarily limited in scope by the size and speed of memory\"\n6. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-6)**\n   [Koskenniemi, Kimmo](https:\/\/en.wikipedia.org\/wiki\/Kimmo_Koskenniemi \"Kimmo Koskenniemi\") (1983), [*Two-level morphology: A general computational model of word-form recognition and production*](https:\/\/web.archive.org\/web\/20181221032913\/http:\/\/www.ling.helsinki.fi\/~koskenni\/doc\/Two-LevelMorphology.pdf) (PDF), Department of General Linguistics, [University of Helsinki](https:\/\/en.wikipedia.org\/wiki\/University_of_Helsinki \"University of Helsinki\"), archived from [the original](http:\/\/www.ling.helsinki.fi\/~koskenni\/doc\/Two-LevelMorphology.pdf) (PDF) on 2018-12-21, retrieved 2020-08-20\n7. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-7)** Joshi, A. K., & Weinstein, S. (1981, August). [Control of Inference: Role of Some Aspects of Discourse Structure-Centering](https:\/\/www.ijcai.org\/Proceedings\/81-1\/Papers\/071.pdf). In *IJCAI* (pp. 385–387).\n8. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-8)**\n   Guida, G.; Mauri, G. (July 1986). \"Evaluation of natural language processing systems: Issues and approaches\". *Proceedings of the IEEE*. **74** (7): 1026–1035\\. [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[1986IEEEP..74.1026G](https:\/\/ui.adsabs.harvard.edu\/abs\/1986IEEEP..74.1026G). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1109\/PROC.1986.13580](https:\/\/doi.org\/10.1109%2FPROC.1986.13580). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1558-2256](https:\/\/search.worldcat.org\/issn\/1558-2256). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [30688575](https:\/\/api.semanticscholar.org\/CorpusID:30688575).\n9. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-9)** Chomskyan linguistics encourages the investigation of \"[corner cases](https:\/\/en.wikipedia.org\/wiki\/Corner_case \"Corner case\")\" that stress the limits of its theoretical models (comparable to [pathological](https:\/\/en.wikipedia.org\/wiki\/Pathological_\\(mathematics\\) \"Pathological (mathematics)\") phenomena in mathematics), typically created using [thought experiments](https:\/\/en.wikipedia.org\/wiki\/Thought_experiment \"Thought experiment\"), rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in [corpus linguistics](https:\/\/en.wikipedia.org\/wiki\/Corpus_linguistics \"Corpus linguistics\"). The creation and use of such [corpora](https:\/\/en.wikipedia.org\/wiki\/Text_corpus \"Text corpus\") of real-world data is a fundamental part of machine-learning algorithms for natural language processing. In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called \"[poverty of the stimulus](https:\/\/en.wikipedia.org\/wiki\/Poverty_of_the_stimulus \"Poverty of the stimulus\")\" argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing. As a result, the Chomskyan paradigm discouraged the application of such models to language processing.\n10. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-10)**\n    Bengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003). [\"A neural probabilistic language model\"](https:\/\/dl.acm.org\/doi\/10.5555\/944919.944966). *The Journal of Machine Learning Research*. **3**: 1137–1155 – via ACM Digital Library.\n11. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-11)**\n    Mikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010). [\"Recurrent neural network based language model\"](https:\/\/gwern.net\/doc\/ai\/nn\/rnn\/2010-mikolov.pdf) (PDF). *Interspeech 2010*. pp. 1045–1048\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.21437\/Interspeech.2010-343](https:\/\/doi.org\/10.21437%2FInterspeech.2010-343). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [17048224](https:\/\/api.semanticscholar.org\/CorpusID:17048224).\n12. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-goldberg:nnlp17_12-0)**\n    Goldberg, Yoav (2016). [\"A Primer on Neural Network Models for Natural Language Processing\"](https:\/\/doi.org\/10.1613%2Fjair.4992). *Journal of Artificial Intelligence Research*. **57**: 345–420\\. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1807\\.10854](https:\/\/arxiv.org\/abs\/1807.10854). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1613\/jair.4992](https:\/\/doi.org\/10.1613%2Fjair.4992). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [8273530](https:\/\/api.semanticscholar.org\/CorpusID:8273530).\n13. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-goodfellow:book16_13-0)**\n    Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016). [*Deep Learning*](http:\/\/www.deeplearningbook.org\/). MIT Press.\n14. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-jozefowicz:lm16_14-0)**\n    Jozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016). *Exploring the Limits of Language Modeling*. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1602\\.02410](https:\/\/arxiv.org\/abs\/1602.02410). [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[2016arXiv160202410J](https:\/\/ui.adsabs.harvard.edu\/abs\/2016arXiv160202410J).\n15. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-choe:emnlp16_15-0)**\n    Choe, Do Kook; Charniak, Eugene. [\"Parsing as Language Modeling\"](https:\/\/web.archive.org\/web\/20181023034804\/https:\/\/aclanthology.coli.uni-saarland.de\/papers\/D16-1257\/d16-1257). *Emnlp 2016*. Archived from [the original](https:\/\/aclanthology.coli.uni-saarland.de\/papers\/D16-1257\/d16-1257) on 2018-10-23. Retrieved 2018-10-22.\n16. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-vinyals:nips15_16-0)**\n    Vinyals, Oriol; et al. (2014). [\"Grammar as a Foreign Language\"](https:\/\/papers.nips.cc\/paper\/5635-grammar-as-a-foreign-language.pdf) (PDF). *Nips2015*. [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[1412\\.7449](https:\/\/arxiv.org\/abs\/1412.7449). [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[2014arXiv1412.7449V](https:\/\/ui.adsabs.harvard.edu\/abs\/2014arXiv1412.7449V).\n17. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-17)**\n    Turchin, Alexander; Florez Builes, Luisa F. (2021-03-19). [\"Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC8120048). *Journal of Diabetes Science and Technology*. **15** (3): 553–560\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1177\/19322968211000831](https:\/\/doi.org\/10.1177%2F19322968211000831). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1932-2968](https:\/\/search.worldcat.org\/issn\/1932-2968). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [8120048](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC8120048). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [33736486](https:\/\/pubmed.ncbi.nlm.nih.gov\/33736486).\n18. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-18)**\n    Lee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10). [\"Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9233261). *JMIR Medical Informatics*. **10** (6) e38482. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.2196\/38482](https:\/\/doi.org\/10.2196%2F38482). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [2291-9694](https:\/\/search.worldcat.org\/issn\/2291-9694). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [9233261](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9233261). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [35687381](https:\/\/pubmed.ncbi.nlm.nih.gov\/35687381).\n19. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-winograd:shrdlu71_19-0)**\n    Winograd, Terry (1971). [*Procedures as a Representation for Data in a Computer Program for Understanding Natural Language*](http:\/\/hci.stanford.edu\/winograd\/shrdlu\/) (Thesis).\n20. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-schank77_20-0)**\n    Schank, Roger C.; Abelson, Robert P. (1977). *Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures*. Hillsdale: Erlbaum. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [0-470-99033-3](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/0-470-99033-3 \"Special:BookSources\/0-470-99033-3\")\n    \n    .\n21. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-johnson:eacl:ilcl09_21-0)** [Mark Johnson. How the statistical revolution changes (computational) linguistics.](http:\/\/www.aclweb.org\/anthology\/W09-0103) Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics.\n22. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-resnik:langlog11_22-0)** [Philip Resnik. Four revolutions.](http:\/\/languagelog.ldc.upenn.edu\/nll\/?p=2946) Language Log, February 5, 2011.\n23. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-23)**\n    Socher, Richard. [\"Deep Learning For NLP-ACL 2012 Tutorial\"](https:\/\/web.archive.org\/web\/20210414054126\/https:\/\/www.socher.org\/index.php\/Main\/DeepLearningForNLP-ACL2012Tutorial). *www.socher.org*. Archived from [the original](https:\/\/www.socher.org\/index.php\/Main\/DeepLearningForNLP-ACL2012Tutorial) on 2021-04-14. Retrieved 2020-08-17.\n    \n    This was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. \\[Link is broken, try <http:\/\/web.stanford.edu\/class\/cs224n\/>\\]\n24. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-24)**\n    Segev, Elad (2022). [*Semantic Network Analysis in Social Sciences*](https:\/\/www.routledge.com\/Semantic-Network-Analysis-in-Social-Sciences\/Segev\/p\/book\/9780367636524). London: Routledge. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-367-63652-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-367-63652-4 \"Special:BookSources\/978-0-367-63652-4\")\n    \n    . [Archived](https:\/\/web.archive.org\/web\/20211205140726\/https:\/\/www.routledge.com\/Semantic-Network-Analysis-in-Social-Sciences\/Segev\/p\/book\/9780367636524) from the original on 5 December 2021. Retrieved 5 December 2021.\n25. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-25)**\n    Yi, Chucai; [Tian, Yingli](https:\/\/en.wikipedia.org\/wiki\/Yingli_Tian \"Yingli Tian\") (2012), \"Assistive Text Reading from Complex Background for Blind Persons\", *Camera-Based Document Analysis and Recognition*, Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. 15–28, [CiteSeerX](https:\/\/en.wikipedia.org\/wiki\/CiteSeerX_\\(identifier\\) \"CiteSeerX (identifier)\") [10\\.1.1.668.869](https:\/\/citeseerx.ist.psu.edu\/viewdoc\/summary?doi=10.1.1.668.869), [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1007\/978-3-642-29364-1\\_2](https:\/\/doi.org\/10.1007%2F978-3-642-29364-1_2), [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-3-642-29363-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-3-642-29363-4 \"Special:BookSources\/978-3-642-29363-4\")\n    \n    `{{citation}}`: CS1 maint: work parameter with ISBN ([link](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_work_parameter_with_ISBN \"Category:CS1 maint: work parameter with ISBN\"))\n26. ^ [***a***](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-:0_26-0) [***b***](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-:0_26-1)\n    [\"Natural Language Processing (NLP) - A Complete Guide\"](https:\/\/www.deeplearning.ai\/resources\/natural-language-processing\/). *www.deeplearning.ai*. 2023-01-11. Retrieved 2024-05-05.\n27. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-27)**\n    [\"GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks\"](https:\/\/www.geeksforgeeks.org\/nlp\/tokenization-in-natural-language-processing-nlp\/). *Geeksforgeeks*. 25 June 2024.\n28. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-28)**\n    Torabi, Mohhamadreza; Ghasemi, Fahimeh. [\"Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC12391535). *Scientific Reports*. **1** (1): 1–15\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1038\/s41598-025-14699-1](https:\/\/doi.org\/10.1038%2Fs41598-025-14699-1). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [12391535](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC12391535).\n29. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-29)**\n    [\"What is Natural Language Processing? Intro to NLP in Machine Learning\"](https:\/\/www.gyansetu.in\/what-is-natural-language-processing\/). *GyanSetu\\!*. 2020-12-06. Retrieved 2021-01-09.\n30. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-30)**\n    Kishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012). [\"Manipuri Morpheme Identification\"](http:\/\/aclweb.org\/anthology\/\/W\/W12\/W12-5008.pdf) (PDF). *Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP)*. COLING 2012, Mumbai, December 2012: 95–108\\.\n    \n    `{{cite journal}}`: CS1 maint: location ([link](https:\/\/en.wikipedia.org\/wiki\/Category:CS1_maint:_location \"Category:CS1 maint: location\"))\n31. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-31)**\n    Klein, Dan; Manning, Christopher D. (2002). [\"Natural language grammar induction using a constituent-context model\"](http:\/\/papers.nips.cc\/paper\/1945-natural-language-grammar-induction-using-a-constituent-context-model.pdf) (PDF). *Advances in Neural Information Processing Systems*.\n32. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-32)**\n    Kariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023). [\"Precision information extraction for rare disease epidemiology at scale\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9972634). *Journal of Translational Medicine*. **21** (1): 157. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1186\/s12967-023-04011-y](https:\/\/doi.org\/10.1186%2Fs12967-023-04011-y). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [9972634](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC9972634). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [36855134](https:\/\/pubmed.ncbi.nlm.nih.gov\/36855134).\n33. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-rte:11_33-0)** PASCAL Recognizing Textual Entailment Challenge (RTE-7) <https:\/\/tac.nist.gov\/\/2011\/RTE\/>\n34. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-34)**\n    Lippi, Marco; Torroni, Paolo (2016-04-20). [\"Argumentation Mining: State of the Art and Emerging Trends\"](https:\/\/dl.acm.org\/doi\/10.1145\/2850417). *ACM Transactions on Internet Technology*. **16** (2): 1–25\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1145\/2850417](https:\/\/doi.org\/10.1145%2F2850417). [hdl](https:\/\/en.wikipedia.org\/wiki\/Hdl_\\(identifier\\) \"Hdl (identifier)\"):[11585\/523460](https:\/\/hdl.handle.net\/11585%2F523460). [ISSN](https:\/\/en.wikipedia.org\/wiki\/ISSN_\\(identifier\\) \"ISSN (identifier)\") [1533-5399](https:\/\/search.worldcat.org\/issn\/1533-5399). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [9561587](https:\/\/api.semanticscholar.org\/CorpusID:9561587).\n35. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-35)**\n    [\"Argument Mining – IJCAI2016 Tutorial\"](https:\/\/web.archive.org\/web\/20210418083659\/https:\/\/www.i3s.unice.fr\/~villata\/tutorialIJCAI2016.html). *www.i3s.unice.fr*. Archived from [the original](https:\/\/www.i3s.unice.fr\/~villata\/tutorialIJCAI2016.html) on 2021-04-18. Retrieved 2021-03-09.\n36. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-36)**\n    [\"NLP Approaches to Computational Argumentation – ACL 2016, Berlin\"](http:\/\/acl2016tutorial.arg.tech\/). Retrieved 2021-03-09.\n37. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-37)**\n    Administration. [\"Centre for Language Technology (CLT)\"](https:\/\/www.mq.edu.au\/research\/research-centres-groups-and-facilities\/innovative-technologies\/centres\/centre-for-language-technology-clt). *Macquarie University*. Retrieved 2021-01-11.\n38. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-38)**\n    [\"Shared Task: Grammatical Error Correction\"](https:\/\/www.comp.nus.edu.sg\/~nlp\/conll13st.html). *www.comp.nus.edu.sg*. Retrieved 2021-01-11.\n39. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-39)**\n    [\"Shared Task: Grammatical Error Correction\"](https:\/\/www.comp.nus.edu.sg\/~nlp\/conll14st.html). *www.comp.nus.edu.sg*. Retrieved 2021-01-11.\n40. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-40)**\n    Duan, Yucong; Cruz, Christophe (2011). [\"Formalizing Semantic of Natural Language through Conceptualization from Existence\"](https:\/\/web.archive.org\/web\/20111009135952\/http:\/\/www.ijimt.org\/abstract\/100-E00187.htm). *International Journal of Innovation, Management and Technology*. **2** (1): 37–42\\. Archived from [the original](http:\/\/www.ijimt.org\/abstract\/100-E00187.htm) on 2011-10-09.\n41. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-41)**\n    [\"U B U W E B :: Racter\"](http:\/\/www.ubu.com\/historical\/racter\/index.html). *www.ubu.com*. Retrieved 2020-08-17.\n42. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-42)**\n    Writer, Beta (2019). *Lithium-Ion Batteries*. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1007\/978-3-030-16800-1](https:\/\/doi.org\/10.1007%2F978-3-030-16800-1). [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-3-030-16799-8](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-3-030-16799-8 \"Special:BookSources\/978-3-030-16799-8\")\n    \n    . [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [155818532](https:\/\/api.semanticscholar.org\/CorpusID:155818532).\n43. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-43)**\n    [\"Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube\"](https:\/\/ghostarchive.org\/varchive\/youtube\/20211030\/7dtl650D0y0). *www.youtube.com*. 11 April 2019. Archived from [the original](https:\/\/www.youtube.com\/watch?v=7dtl650D0y0) on 2021-10-30. Retrieved 2021-01-11.\n44. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-44)**\n    Robertson, Adi (2022-04-06). [\"OpenAI's DALL-E AI image generator can now edit pictures, too\"](https:\/\/www.theverge.com\/2022\/4\/6\/23012123\/openai-clip-dalle-2-ai-text-to-image-generator-testing). *The Verge*. Retrieved 2022-06-07.\n45. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-45)**\n    [\"The Stanford Natural Language Processing Group\"](https:\/\/nlp.stanford.edu\/projects\/text2scene.shtml). *nlp.stanford.edu*. Retrieved 2022-06-07.\n46. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-46)**\n    Coyne, Bob; Sproat, Richard (2001-08-01). [\"WordsEye\"](https:\/\/doi.org\/10.1145\/383259.383316). *Proceedings of the 28th annual conference on Computer graphics and interactive techniques*. SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. 487–496\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1145\/383259.383316](https:\/\/doi.org\/10.1145%2F383259.383316). [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-1-58113-374-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1-58113-374-5 \"Special:BookSources\/978-1-58113-374-5\")\n    \n    . [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [3842372](https:\/\/api.semanticscholar.org\/CorpusID:3842372).\n47. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-47)**\n    [\"Google announces AI advances in text-to-video, language translation, more\"](https:\/\/venturebeat.com\/ai\/google-announces-ai-advances-in-text-to-video-language-translation-more\/). *VentureBeat*. 2022-11-02. Retrieved 2022-11-09.\n48. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-48)**\n    Vincent, James (2022-09-29). [\"Meta's new text-to-video AI generator is like DALL-E for video\"](https:\/\/www.theverge.com\/2022\/9\/29\/23378210\/meta-text-to-video-ai-generation-make-a-video-model-dall-e). *The Verge*. Retrieved 2022-11-09.\n49. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-49)**\n    [\"Previous shared tasks \\| CoNLL\"](https:\/\/www.conll.org\/previous-tasks). *www.conll.org*. Retrieved 2021-01-11.\n50. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-50)**\n    [\"Cognition\"](https:\/\/web.archive.org\/web\/20200715113427\/https:\/\/www.lexico.com\/definition\/cognition). *Lexico*. [Oxford University Press](https:\/\/en.wikipedia.org\/wiki\/Oxford_University_Press \"Oxford University Press\") and [Dictionary.com](https:\/\/en.wikipedia.org\/wiki\/Dictionary.com \"Dictionary.com\"). Archived from [the original](https:\/\/www.lexico.com\/definition\/cognition) on July 15, 2020. Retrieved 6 May 2020.\n51. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-51)**\n    [\"Ask the Cognitive Scientist\"](http:\/\/www.aft.org\/newspubs\/periodicals\/ae\/summer2002\/willingham.cfm). *American Federation of Teachers*. 8 August 2014. \"Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind.\"\n52. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-52)**\n    Robinson, Peter (2008). *Handbook of Cognitive Linguistics and Second Language Acquisition*. Routledge. pp. 3–8\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-805-85352-0](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-805-85352-0 \"Special:BookSources\/978-0-805-85352-0\")\n    \n    .\n53. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-53)**\n    Lakoff, George (1999). *Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm*. New York Basic Books. pp. 569–583\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-465-05674-3](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-465-05674-3 \"Special:BookSources\/978-0-465-05674-3\")\n    \n    .\n54. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-54)**\n    Strauss, Claudia (1999). *A Cognitive Theory of Cultural Meaning*. Cambridge University Press. pp. 156–164\\. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-521-59541-4](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-521-59541-4 \"Special:BookSources\/978-0-521-59541-4\")\n    \n    .\n55. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-55)** [US patent 9269353](https:\/\/worldwide.espacenet.com\/textdoc?DB=EPODOC&IDX=US9269353)\n56. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-56)**\n    [\"Universal Conceptual Cognitive Annotation (UCCA)\"](https:\/\/universalconceptualcognitiveannotation.github.io\/). *Universal Conceptual Cognitive Annotation (UCCA)*. Retrieved 2021-01-11.\n57. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-57)** Rodríguez, F. C., & Mairal-Usón, R. (2016). [Building an RRG computational grammar](https:\/\/www.redalyc.org\/pdf\/1345\/134549291020.pdf). *Onomazein*, (34), 86–117.\n58. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-58)**\n    [\"Fluid Construction Grammar – A fully operational processing system for construction grammars\"](https:\/\/www.fcg-net.org\/). Retrieved 2021-01-11.\n59. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-59)**\n    [\"ACL Member Portal \\| The Association for Computational Linguistics Member Portal\"](https:\/\/www.aclweb.org\/portal\/). *www.aclweb.org*. Retrieved 2021-01-11.\n60. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-60)**\n    [\"Chunks and Rules\"](https:\/\/www.w3.org\/Data\/demos\/chunks\/chunks.html). *W3C*. Retrieved 2021-01-11.\n61. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-61)**\n    Socher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014). [\"Grounded Compositional Semantics for Finding and Describing Images with Sentences\"](https:\/\/doi.org\/10.1162%2Ftacl_a_00177). *Transactions of the Association for Computational Linguistics*. **2**: 207–218\\. [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1162\/tacl\\_a\\_00177](https:\/\/doi.org\/10.1162%2Ftacl_a_00177). [S2CID](https:\/\/en.wikipedia.org\/wiki\/S2CID_\\(identifier\\) \"S2CID (identifier)\") [2317858](https:\/\/api.semanticscholar.org\/CorpusID:2317858).\n62. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-62)**\n    Dasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). \"Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al\". [arXiv](https:\/\/en.wikipedia.org\/wiki\/ArXiv_\\(identifier\\) \"ArXiv (identifier)\"):[2207\\.07051](https:\/\/arxiv.org\/abs\/2207.07051) \\[[cs.CL](https:\/\/arxiv.org\/archive\/cs.CL)\\].\n63. **[^](https:\/\/en.wikipedia.org\/wiki\/Natural_language_processing#cite_ref-63)**\n    Friston, Karl J. (2022). *Active Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference*. The MIT Press. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n    \n    [978-0-262-36997-8](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-36997-8 \"Special:BookSources\/978-0-262-36997-8\")\n    \n    .\n\n- Bates, M (1995). [\"Models of natural language understanding\"](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC40721). *Proceedings of the National Academy of Sciences of the United States of America*. **92** (22): 9977–9982\\. [Bibcode](https:\/\/en.wikipedia.org\/wiki\/Bibcode_\\(identifier\\) \"Bibcode (identifier)\"):[1995PNAS...92.9977B](https:\/\/ui.adsabs.harvard.edu\/abs\/1995PNAS...92.9977B). [doi](https:\/\/en.wikipedia.org\/wiki\/Doi_\\(identifier\\) \"Doi (identifier)\"):[10\\.1073\/pnas.92.22.9977](https:\/\/doi.org\/10.1073%2Fpnas.92.22.9977). [PMC](https:\/\/en.wikipedia.org\/wiki\/PMC_\\(identifier\\) \"PMC (identifier)\") [40721](https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC40721). [PMID](https:\/\/en.wikipedia.org\/wiki\/PMID_\\(identifier\\) \"PMID (identifier)\") [7479812](https:\/\/pubmed.ncbi.nlm.nih.gov\/7479812).\n- Steven Bird, Ewan Klein, and Edward Loper (2009). *Natural Language Processing with Python*. O'Reilly Media. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-596-51649-9](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-596-51649-9 \"Special:BookSources\/978-0-596-51649-9\")\n  .\n- [Kenna Hughes-Castleberry](https:\/\/en.wikipedia.org\/w\/index.php?title=Kenna_Hughes-Castleberry&action=edit&redlink=1 \"Kenna Hughes-Castleberry (page does not exist)\"), \"A Murder Mystery Puzzle: The literary puzzle *[Cain's Jawbone](https:\/\/en.wikipedia.org\/wiki\/Cain%27s_Jawbone \"Cain's Jawbone\")*, which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms\", *[Scientific American](https:\/\/en.wikipedia.org\/wiki\/Scientific_American \"Scientific American\")*, vol. 329, no. 4 (November 2023), pp. 81–82. \"This murder mystery competition has revealed that although NLP ([natural-language processing](https:\/\/en.wikipedia.org\/wiki\/Natural-language_processing \"Natural-language processing\")) models are capable of incredible feats, their abilities are very much limited by the amount of [context](https:\/\/en.wikipedia.org\/wiki\/Context_\\(linguistics\\) \"Context (linguistics)\") they receive. This \\[...\\] could cause \\[difficulties\\] for researchers who hope to use them to do things such as analyze [ancient languages](https:\/\/en.wikipedia.org\/wiki\/Ancient_language \"Ancient language\"). In some cases, there are few historical records on long-gone [civilizations](https:\/\/en.wikipedia.org\/wiki\/Civilization \"Civilization\") to serve as [training data](https:\/\/en.wikipedia.org\/wiki\/Training_data \"Training data\") for such a purpose.\" (p. 82.)\n- Daniel Jurafsky and James H. Martin (2008). *Speech and Language Processing*, 2nd edition. Pearson Prentice Hall. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-13-187321-6](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-13-187321-6 \"Special:BookSources\/978-0-13-187321-6\")\n  .\n- Mohamed Zakaria Kurdi (2016). *Natural Language Processing and Computational Linguistics: speech, morphology, and syntax*, Volume 1. ISTE-Wiley. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-1848218482](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1848218482 \"Special:BookSources\/978-1848218482\")\n  .\n- Mohamed Zakaria Kurdi (2017). *Natural Language Processing and Computational Linguistics: semantics, discourse, and applications*, Volume 2. ISTE-Wiley. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-1848219212](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-1848219212 \"Special:BookSources\/978-1848219212\")\n  .\n- Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008). *Introduction to Information Retrieval*. Cambridge University Press. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-521-86571-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-521-86571-5 \"Special:BookSources\/978-0-521-86571-5\")\n  . [Official html and pdf versions available without charge.](http:\/\/nlp.stanford.edu\/IR-book\/)\n- Christopher D. Manning and Hinrich Schütze (1999). *Foundations of Statistical Natural Language Processing*. The MIT Press. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-262-13360-9](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-262-13360-9 \"Special:BookSources\/978-0-262-13360-9\")\n  .\n- David M. W. Powers and Christopher C. R. Turk (1989). *Machine Learning of Natural Language*. Springer-Verlag. [ISBN](https:\/\/en.wikipedia.org\/wiki\/ISBN_\\(identifier\\) \"ISBN (identifier)\")\n  [978-0-387-19557-5](https:\/\/en.wikipedia.org\/wiki\/Special:BookSources\/978-0-387-19557-5 \"Special:BookSources\/978-0-387-19557-5\")\n  .\n\n- [![Wikimedia Commons logo](https:\/\/upload.wikimedia.org\/wikipedia\/en\/thumb\/4\/4a\/Commons-logo.svg\/20px-Commons-logo.svg.png)](https:\/\/en.wikipedia.org\/wiki\/File:Commons-logo.svg) Media related to [Natural language processing](https:\/\/commons.wikimedia.org\/wiki\/Category:Natural_language_processing \"commons:Category:Natural language processing\") at Wikimedia Commons","meta_canonical":null}

3. Robots.txt Check

Query:

Response:

4. Spam/Ban Check

Query:

Response:

5. Seen Status Check

ℹ️ Skipped - page is already crawled

📄

INDEXABLE

✅

CRAWLED

1 day ago

🤖

ROBOTS ALLOWED

Page Info Filters

Filter	Status	Condition	Details
HTTP status	PASS	`download_http_code = 200`	HTTP 200
Age cutoff	PASS	`download_stamp > now() - 6 MONTH`	0.1 months ago (distributed domain, exempt)
History drop	PASS	`isNull(history_drop_reason)`	No drop reason
Spam/ban	PASS	`fh_dont_index != 1 AND ml_spam_score = 0`	ml_spam_score=0
Canonical	PASS	`meta_canonical IS NULL OR = '' OR = src_unparsed`	Not set

Page Details

Property	Value
URL	https://en.wikipedia.org/wiki/Natural_language_processing
Last Crawled	2026-04-10 08:13:55 (1 day ago)
First Indexed	2013-08-08 16:21:52 (12 years ago)
HTTP Status Code	200
Meta Title	Natural language processing - Wikipedia
Meta Description	null
Meta Canonical	null
Boilerpipe Text	From Wikipedia, the free encyclopedia Natural language processing ( NLP ) is the processing of natural language information by a computer . NLP is a subfield of computer science and is closely associated with artificial intelligence . NLP is also related to information retrieval , knowledge representation , computational linguistics , and linguistics more broadly. [ 1 ] Major processing tasks in an NLP system include: speech recognition , text classification , natural language understanding , and natural language generation . Natural language processing has its roots in the 1950s. [ 2 ] Already in 1950, Alan Turing published an article titled " Computing Machinery and Intelligence " which proposed what is now called the Turing test as a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language. Symbolic NLP (1950s – early 1990s) [ edit ] A document parsed into an abstract syntax tree The premise of symbolic NLP is often illustrated using John Searle's Chinese room thought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts. 1950s : The Georgetown experiment in 1954 involved fully automatic translation of more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem. [ 3 ] However, real progress was much slower, and after the ALPAC report in 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe [ 4 ] ) until the late 1980s when the first statistical machine translation systems were developed. 1960s : Some notably successful natural language processing systems developed in the 1960s were SHRDLU , a natural language system working in restricted " blocks worlds " with restricted vocabularies, and ELIZA , a simulation of a Rogerian psychotherapy , written by Joseph Weizenbaum between 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the "patient" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to "My head hurts" with "Why do you say your head hurts?". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only twenty words, because that was all that would fit in a computer memory at the time. [ 5 ] 1970s : During the 1970s, many programmers began to write "conceptual ontologies ", which structured real-world information into computer-understandable data. Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981). During this time, the first chatterbots were written (e.g., PARRY ). 1980s : The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of HPSG as a computational operationalization of generative grammar ), morphology (e.g., two-level morphology [ 6 ] ), semantics (e.g., Lesk algorithm ), reference (e.g., within Centering Theory [ 7 ] ) and other areas of natural language understanding (e.g., in the Rhetorical Structure Theory ). Other lines of research were continued, e.g., the development of chatterbots with Racter and Jabberwacky . An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period. [ 8 ] Statistical NLP (1990s–present) [ edit ] Up until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of machine learning algorithms for language processing. This shift was influenced by increasing computational power (see Moore's law ) and a decline in the dominance of Chomskyan linguistic theories... (e.g. transformational grammar ), whose theoretical underpinnings discouraged the sort of corpus linguistics that underlies the machine-learning approach to language processing. [ 9 ] 1990s : Many of the notable early successes in statistical methods in NLP occurred in the field of machine translation , due especially to work at IBM Research, such as IBM alignment models . These systems were able to take advantage of existing multilingual textual corpora that had been produced by the Parliament of Canada and the European Union as a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data. 2000s : With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on unsupervised and semi-supervised learning algorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than supervised learning , and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the World Wide Web ), which can often make up for the worse efficiency if the algorithm used has a low enough time complexity to be practical. 2003: word n-gram model , at the time the best statistical algorithm, is outperformed by a multi-layer perceptron (with a single hidden layer and context length of several words, trained on up to 14 million words, by Bengio et al.) [ 10 ] 2010: Tomáš Mikolov (then a PhD student at Brno University of Technology ) with co-authors applied a simple recurrent neural network with a single hidden layer to language modeling, [ 11 ] and in the following years he went on to develop Word2vec . In the 2010s, representation learning and deep neural network -style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques [ 12 ] [ 13 ] can achieve state-of-the-art results in many natural language tasks, e.g., in language modeling [ 14 ] and parsing. [ 15 ] [ 16 ] This is increasingly important in medicine and healthcare , where NLP helps analyze notes and text in electronic health records that would otherwise be inaccessible for study when seeking to improve care [ 17 ] or protect patient privacy. [ 18 ] Approaches: Symbolic, statistical, neural networks [ edit ] Symbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular: [ 19 ] [ 20 ] such as by writing grammars or devising heuristic rules for stemming . Machine learning approaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach: both statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally. language models , produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce. the larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to intractability problems. Rule-based systems are commonly used: when the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the Apertium system, for preprocessing in NLP pipelines, e.g., tokenization , or for post-processing and transforming the output of NLP pipelines, e.g., for knowledge extraction from syntactic parses. Statistical approach [ edit ] In the late 1980s and mid-1990s, the statistical approach ended a period of AI winter , which was caused by the inefficiencies of the rule-based approaches. [ 21 ] [ 22 ] The earliest decision trees , producing systems of hard if–then rules , were still very similar to the old rule-based approaches. Only the introduction of hidden Markov models , applied to part-of-speech tagging, announced the end of the old rule-based approach. A major drawback of statistical methods is that they require elaborate feature engineering . Since 2015, [ 23 ] neural network –based methods have increasingly replaced traditional statistical approaches, using semantic networks [ 24 ] and word embeddings to capture semantic properties of words. Intermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore. Neural machine translation , based on then-newly invented sequence-to-sequence transformations, made obsolete the intermediate steps, such as word alignment, previously necessary for statistical machine translation . The following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks. Though natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below. Text and speech processing [ edit ] Word cloud of stop words in Hebrew Optical character recognition (OCR) Given an image representing printed text, determine the corresponding text. Speech recognition Given a sound clip of a person or people speaking, determine the textual representation of the speech. This is the opposite of text to speech and is one of the extremely difficult problems colloquially termed " AI-complete " (see above). In natural speech there are hardly any pauses between successive words, and thus speech segmentation is a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed coarticulation , so the conversion of the analog signal to discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent. Speech segmentation Given a sound clip of a person or people speaking, separate it into words. A subtask of speech recognition and typically grouped with it. Text-to-speech Given a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired. [ 25 ] Word segmentation ( Tokenization ) Tokenization is a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods. [ 26 ] For a language like English , this is fairly trivial, since words are usually separated by spaces. However, some written languages like Chinese , Japanese and Thai do not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the vocabulary and morphology of words in the language. Sometimes this process is also used in cases like bag of words (BOW) creation in data mining. [ 27 ] [ 28 ] Morphological analysis [ edit ] Lemmatization of Basque words Lemmatization The task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form. [ 29 ] Morphological segmentation Separate words into individual morphemes and identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the morphology ( i.e. , the structure of words) of the language being considered. English has fairly simple morphology, especially inflectional morphology , and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., "open, opens, opened, opening") as separate words. In languages such as Turkish or Meitei , a highly agglutinated Indian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms. [ 30 ] Part-of-speech tagging Given a sentence, determine the part of speech (POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, "book" can be a noun ("the book on the table") or verb ("to book a flight"); "set" can be a noun, verb or adjective ; and "out" can be any of at least five different parts of speech. Stemming The process of reducing inflected (or sometimes derived) words to a base form (e.g., "close" will be the root for "closed", "closing", "close", "closer" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary. Grammar induction [ 31 ] Generate a formal grammar that describes a language's syntax. Sentence breaking (also known as " sentence boundary disambiguation ") Given a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by periods or other punctuation marks , but these same characters can serve other purposes (e.g., marking abbreviations ). Parsing Determine the parse tree (grammatical analysis) of a given sentence. The grammar for natural languages is ambiguous and typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing: dependency parsing and constituency parsing . Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a probabilistic context-free grammar (PCFG) (see also stochastic grammar ). Lexical semantics (of individual words in context) [ edit ] An entity linking pipeline Lexical semantics What is the computational meaning of individual words in context? Distributional semantics How can we learn semantic representations from data? Named entity recognition (NER) Given a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although capitalization can aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of named entity , and in any case, is often inaccurate or insufficient. For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized. Furthermore, many other languages in non-Western scripts (e.g. Chinese or Arabic ) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example, German capitalizes all nouns , regardless of whether they are names, and French and Spanish do not capitalize names that serve as adjectives . This task is also referred to as token classification. [ 32 ] Sentiment analysis (see also Multimodal sentiment analysis ) Sentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as word n-grams , Term Frequency-Inverse Document Frequency (TF-IDF) features, hand-generated features, or employ deep learning models designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms. [ 26 ] Terminology extraction The goal of terminology extraction is to automatically extract relevant terms from a given corpus. Word-sense disambiguation (WSD) Many words have more than one meaning ; we have to select the meaning which makes the most sense in context. For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as WordNet . Entity linking Many words—typically proper names—refer to named entities ; here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context. Relational semantics (semantics of individual sentences) [ edit ] Relationship extraction Given a chunk of text, identify the relationships among named entities (e.g. who is married to whom). Semantic parsing Given a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in AMR parsing ) or in accordance with a logical formalism (e.g., in DRT parsing ). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see Natural language understanding below). Semantic role labelling (see also implicit semantic role labelling below) Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal frames ), then identify and classify the frame elements ( semantic roles ). Discourse (semantics beyond individual sentences) [ edit ] Coreference resolution Given a sentence or larger chunk of text, determine which words ("mentions") refer to the same objects ("entities"). Anaphora resolution is a specific example of this task, and is specifically concerned with matching up pronouns with the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called "bridging relationships" involving referring expressions . For example, in a sentence such as "He entered John's house through the front door", "the front door" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to). Discourse analysis This rubric includes several related tasks. One task is discourse parsing, i.e., identifying the discourse structure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast). Another possible task is recognizing and classifying the speech acts in a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.). Implicit semantic role labelling Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal frames ) and their explicit semantic roles in the current sentence (see Semantic role labelling above). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to pro-drop languages . Recognizing textual entailment Given two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false. [ 33 ] Topic segmentation and recognition Given a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment. Argument mining The goal of argument mining is the automatic extraction and identification of argumentative structures from natural language text with the aid of computer programs. [ 34 ] Such argumentative structures include the premise, conclusions, the argument scheme and the relationship between the main and subsidiary argument, or the main and counter-argument within discourse. [ 35 ] [ 36 ] Higher-level NLP applications [ edit ] Machine translation in Firefox Automatic summarization (text summarization) Produce a readable summary of a chunk of text. Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper. Grammatical error correction Grammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011. [ 37 ] [ 38 ] [ 39 ] As far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as GPT-2 , this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications. Logic translation Translate a text from a natural language into formal logic. Machine translation (MT) Automatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed " AI-complete ", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly. Natural language understanding (NLU) Convert chunks of text into more formal representations such as first-order logic structures that are easier for computer programs to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as closed-world assumption (CWA) vs. open-world assumption , or subjective Yes/No vs. objective True/False is expected for the construction of a basis of semantics formalization. [ 40 ] Natural language generation (NLG): Convert information from computer databases or semantic intents into readable human language. Book generation Not an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter, The policeman's beard is half-constructed ). [ 41 ] The first published work by a neural network was published in 2018, 1 the Road , marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free) language models . The first machine-generated science book was published in 2019 (Beta Writer, Lithium-Ion Batteries , Springer, Cham). [ 42 ] Unlike Racter and 1 the Road , this is grounded on factual knowledge and based on text summarization. Document AI A Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants. [ 43 ] Dialogue management Computer systems intended to converse with a human. Question answering Given a human-language question, determine its answer. Typical questions have a specific right answer (such as "What is the capital of Canada?"), but sometimes open-ended questions are also considered (such as "What is the meaning of life?"). Text-to-image generation Given a description of an image, generate an image that matches the description. [ 44 ] Text-to-scene generation Given a description of a scene, generate a 3D model of the scene. [ 45 ] [ 46 ] Text-to-video Given a description of a video, generate a video that matches the description. [ 47 ] [ 48 ] General tendencies and (possible) future directions [ edit ] Based on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed: [ 49 ] Interest on increasingly abstract, "cognitive" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09/2017–18: dependency syntax, 2004–05/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing). Increasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+/100+ languages) Elimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems) Most higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above). Cognition refers to "the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses." [ 50 ] Cognitive science is the interdisciplinary, scientific study of the mind and its processes. [ 51 ] Cognitive linguistics is an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics. [ 52 ] Especially during the age of symbolic NLP , the area of computational linguistics maintained strong ties with cognitive studies. As an example, George Lakoff offers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics, [ 53 ] with two defining aspects: Apply the theory of conceptual metaphor , explained by Lakoff as "the understanding of one idea, in terms of another" which provides an idea of the intent of the author. [ 54 ] For example, consider the English word big . When used in a comparison ("That is a big tree"), the author's intent is to imply that the tree is physically large relative to other trees or the authors experience. When used metaphorically ("Tomorrow is a big day"), the author's intent to imply importance . The intent behind other usages, like in "She is a big person", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information. Assign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a probabilistic context-free grammar (PCFG). The mathematical equation for such algorithms is presented in US Patent 9269353 : [ 55 ] Where RMM is the relative measure of meaning token is any block of text, sentence, phrase or word N is the number of tokens being analyzed PMM is the probable measure of meaning based on a corpora d is the non zero location of the token along the sequence of N tokens PF is the probability function specific to a language Ties with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar, [ 56 ] functional grammar, [ 57 ] construction grammar, [ 58 ] computational psycholinguistics and cognitive neuroscience (e.g., ACT-R ), however, with limited uptake in mainstream NLP (as measured by presence on major conferences [ 59 ] of the ACL ). More recently, ideas of cognitive NLP have been revived as an approach to achieve explainability , e.g., under the notion of "cognitive AI". [ 60 ] Likewise, ideas of cognitive NLP are inherent to neural models multimodal NLP (although rarely made explicit) [ 61 ] and developments in artificial intelligence , specifically tools and technologies using large language model approaches [ 62 ] and new directions in artificial general intelligence based on the free energy principle [ 63 ] by British neuroscientist and theoretician at University College London Karl J. Friston . 1 the Road Artificial intelligence detection software Automated essay scoring Biomedical text mining Compound term processing Computational linguistics Computer-assisted reviewing Controlled natural language Deep learning Deep linguistic processing Distributional semantics Foreign language reading aid Foreign language writing aid Information extraction Information retrieval Language and Communication Technologies Language model Language technology Latent semantic indexing Multi-agent system Native-language identification Natural-language programming Natural-language understanding Natural-language search Outline of natural language processing Query expansion Query understanding Reification (linguistics) Speech processing Spoken dialogue systems Text-proofing Text simplification Transformer (machine learning model) Truecasing Question answering Word2vec ^ Eisenstein, Jacob (October 1, 2019). Introduction to Natural Language Processing . The MIT Press. p. 1. ISBN 978-0-262-04284-0 . ^ "NLP" . ^ Hutchins, J. (2005). "The history of machine translation in a nutshell" (PDF) . Archived from the original (PDF) on 2019-07-13 . Retrieved 2019-02-04 . [ self-published source ] ^ "ALPAC: the (in)famous report", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12. ^ Crevier 1993 , pp. 146–148 , see also Buchanan 2005 , p. 56 : "Early programs were necessarily limited in scope by the size and speed of memory" ^ Koskenniemi, Kimmo (1983), Two-level morphology: A general computational model of word-form recognition and production (PDF) , Department of General Linguistics, University of Helsinki , archived from the original (PDF) on 2018-12-21 , retrieved 2020-08-20 ^ Joshi, A. K., & Weinstein, S. (1981, August). Control of Inference: Role of Some Aspects of Discourse Structure-Centering . In IJCAI (pp. 385–387). ^ Guida, G.; Mauri, G. (July 1986). "Evaluation of natural language processing systems: Issues and approaches". Proceedings of the IEEE . 74 (7): 1026– 1035. Bibcode : 1986IEEEP..74.1026G . doi : 10.1109/PROC.1986.13580 . ISSN 1558-2256 . S2CID 30688575 . ^ Chomskyan linguistics encourages the investigation of " corner cases " that stress the limits of its theoretical models (comparable to pathological phenomena in mathematics), typically created using thought experiments , rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in corpus linguistics . The creation and use of such corpora of real-world data is a fundamental part of machine-learning algorithms for natural language processing. In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called " poverty of the stimulus " argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing. As a result, the Chomskyan paradigm discouraged the application of such models to language processing. ^ Bengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003). "A neural probabilistic language model" . The Journal of Machine Learning Research . 3 : 1137– 1155 – via ACM Digital Library. ^ Mikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010). "Recurrent neural network based language model" (PDF) . Interspeech 2010 . pp. 1045– 1048. doi : 10.21437/Interspeech.2010-343 . S2CID 17048224 . ^ Goldberg, Yoav (2016). "A Primer on Neural Network Models for Natural Language Processing" . Journal of Artificial Intelligence Research . 57 : 345– 420. arXiv : 1807.10854 . doi : 10.1613/jair.4992 . S2CID 8273530 . ^ Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016). Deep Learning . MIT Press. ^ Jozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016). Exploring the Limits of Language Modeling . arXiv : 1602.02410 . Bibcode : 2016arXiv160202410J . ^ Choe, Do Kook; Charniak, Eugene. "Parsing as Language Modeling" . Emnlp 2016 . Archived from the original on 2018-10-23 . Retrieved 2018-10-22 . ^ Vinyals, Oriol; et al. (2014). "Grammar as a Foreign Language" (PDF) . Nips2015 . arXiv : 1412.7449 . Bibcode : 2014arXiv1412.7449V . ^ Turchin, Alexander; Florez Builes, Luisa F. (2021-03-19). "Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review" . Journal of Diabetes Science and Technology . 15 (3): 553– 560. doi : 10.1177/19322968211000831 . ISSN 1932-2968 . PMC 8120048 . PMID 33736486 . ^ Lee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10). "Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study" . JMIR Medical Informatics . 10 (6) e38482. doi : 10.2196/38482 . ISSN 2291-9694 . PMC 9233261 . PMID 35687381 . ^ Winograd, Terry (1971). Procedures as a Representation for Data in a Computer Program for Understanding Natural Language (Thesis). ^ Schank, Roger C.; Abelson, Robert P. (1977). Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures . Hillsdale: Erlbaum. ISBN 0-470-99033-3 . ^ Mark Johnson. How the statistical revolution changes (computational) linguistics. Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics. ^ Philip Resnik. Four revolutions. Language Log, February 5, 2011. ^ Socher, Richard. "Deep Learning For NLP-ACL 2012 Tutorial" . www.socher.org . Archived from the original on 2021-04-14 . Retrieved 2020-08-17 . This was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. [Link is broken, try http://web.stanford.edu/class/cs224n/ ] ^ Segev, Elad (2022). Semantic Network Analysis in Social Sciences . London: Routledge. ISBN 978-0-367-63652-4 . Archived from the original on 5 December 2021 . Retrieved 5 December 2021 . ^ Yi, Chucai; Tian, Yingli (2012), "Assistive Text Reading from Complex Background for Blind Persons", Camera-Based Document Analysis and Recognition , Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. 15– 28, CiteSeerX 10.1.1.668.869 , doi : 10.1007/978-3-642-29364-1_2 , ISBN 978-3-642-29363-4 {{ citation }} : CS1 maint: work parameter with ISBN ( link ) ^ a b "Natural Language Processing (NLP) - A Complete Guide" . www.deeplearning.ai . 2023-01-11 . Retrieved 2024-05-05 . ^ "GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks" . Geeksforgeeks . 25 June 2024. ^ Torabi, Mohhamadreza; Ghasemi, Fahimeh. "Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh" . Scientific Reports . 1 (1): 1– 15. doi : 10.1038/s41598-025-14699-1 . PMC 12391535 . ^ "What is Natural Language Processing? Intro to NLP in Machine Learning" . GyanSetu! . 2020-12-06 . Retrieved 2021-01-09 . ^ Kishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012). "Manipuri Morpheme Identification" (PDF) . Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP) . COLING 2012, Mumbai, December 2012: 95– 108. {{ cite journal }} : CS1 maint: location ( link ) ^ Klein, Dan; Manning, Christopher D. (2002). "Natural language grammar induction using a constituent-context model" (PDF) . Advances in Neural Information Processing Systems . ^ Kariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023). "Precision information extraction for rare disease epidemiology at scale" . Journal of Translational Medicine . 21 (1): 157. doi : 10.1186/s12967-023-04011-y . PMC 9972634 . PMID 36855134 . ^ PASCAL Recognizing Textual Entailment Challenge (RTE-7) https://tac.nist.gov//2011/RTE/ ^ Lippi, Marco; Torroni, Paolo (2016-04-20). "Argumentation Mining: State of the Art and Emerging Trends" . ACM Transactions on Internet Technology . 16 (2): 1– 25. doi : 10.1145/2850417 . hdl : 11585/523460 . ISSN 1533-5399 . S2CID 9561587 . ^ "Argument Mining – IJCAI2016 Tutorial" . www.i3s.unice.fr . Archived from the original on 2021-04-18 . Retrieved 2021-03-09 . ^ "NLP Approaches to Computational Argumentation – ACL 2016, Berlin" . Retrieved 2021-03-09 . ^ Administration. "Centre for Language Technology (CLT)" . Macquarie University . Retrieved 2021-01-11 . ^ "Shared Task: Grammatical Error Correction" . www.comp.nus.edu.sg . Retrieved 2021-01-11 . ^ "Shared Task: Grammatical Error Correction" . www.comp.nus.edu.sg . Retrieved 2021-01-11 . ^ Duan, Yucong; Cruz, Christophe (2011). "Formalizing Semantic of Natural Language through Conceptualization from Existence" . International Journal of Innovation, Management and Technology . 2 (1): 37– 42. Archived from the original on 2011-10-09. ^ "U B U W E B :: Racter" . www.ubu.com . Retrieved 2020-08-17 . ^ Writer, Beta (2019). Lithium-Ion Batteries . doi : 10.1007/978-3-030-16800-1 . ISBN 978-3-030-16799-8 . S2CID 155818532 . ^ "Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube" . www.youtube.com . 11 April 2019. Archived from the original on 2021-10-30 . Retrieved 2021-01-11 . ^ Robertson, Adi (2022-04-06). "OpenAI's DALL-E AI image generator can now edit pictures, too" . The Verge . Retrieved 2022-06-07 . ^ "The Stanford Natural Language Processing Group" . nlp.stanford.edu . Retrieved 2022-06-07 . ^ Coyne, Bob; Sproat, Richard (2001-08-01). "WordsEye" . Proceedings of the 28th annual conference on Computer graphics and interactive techniques . SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. 487– 496. doi : 10.1145/383259.383316 . ISBN 978-1-58113-374-5 . S2CID 3842372 . ^ "Google announces AI advances in text-to-video, language translation, more" . VentureBeat . 2022-11-02 . Retrieved 2022-11-09 . ^ Vincent, James (2022-09-29). "Meta's new text-to-video AI generator is like DALL-E for video" . The Verge . Retrieved 2022-11-09 . ^ "Previous shared tasks \| CoNLL" . www.conll.org . Retrieved 2021-01-11 . ^ "Cognition" . Lexico . Oxford University Press and Dictionary.com . Archived from the original on July 15, 2020 . Retrieved 6 May 2020 . ^ "Ask the Cognitive Scientist" . American Federation of Teachers . 8 August 2014. Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind. ^ Robinson, Peter (2008). Handbook of Cognitive Linguistics and Second Language Acquisition . Routledge. pp. 3– 8. ISBN 978-0-805-85352-0 . ^ Lakoff, George (1999). Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm . New York Basic Books. pp. 569– 583. ISBN 978-0-465-05674-3 . ^ Strauss, Claudia (1999). A Cognitive Theory of Cultural Meaning . Cambridge University Press. pp. 156– 164. ISBN 978-0-521-59541-4 . ^ US patent 9269353 ^ "Universal Conceptual Cognitive Annotation (UCCA)" . Universal Conceptual Cognitive Annotation (UCCA) . Retrieved 2021-01-11 . ^ Rodríguez, F. C., & Mairal-Usón, R. (2016). Building an RRG computational grammar . Onomazein , (34), 86–117. ^ "Fluid Construction Grammar – A fully operational processing system for construction grammars" . Retrieved 2021-01-11 . ^ "ACL Member Portal \| The Association for Computational Linguistics Member Portal" . www.aclweb.org . Retrieved 2021-01-11 . ^ "Chunks and Rules" . W3C . Retrieved 2021-01-11 . ^ Socher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014). "Grounded Compositional Semantics for Finding and Describing Images with Sentences" . Transactions of the Association for Computational Linguistics . 2 : 207– 218. doi : 10.1162/tacl_a_00177 . S2CID 2317858 . ^ Dasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). "Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al". arXiv : 2207.07051 [ cs.CL ]. ^ Friston, Karl J. (2022). Active Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference . The MIT Press. ISBN 978-0-262-36997-8 . Bates, M (1995). "Models of natural language understanding" . Proceedings of the National Academy of Sciences of the United States of America . 92 (22): 9977– 9982. Bibcode : 1995PNAS...92.9977B . doi : 10.1073/pnas.92.22.9977 . PMC 40721 . PMID 7479812 . Steven Bird, Ewan Klein, and Edward Loper (2009). Natural Language Processing with Python . O'Reilly Media. ISBN 978-0-596-51649-9 . Kenna Hughes-Castleberry , "A Murder Mystery Puzzle: The literary puzzle Cain's Jawbone , which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms", Scientific American , vol. 329, no. 4 (November 2023), pp. 81–82. "This murder mystery competition has revealed that although NLP ( natural-language processing ) models are capable of incredible feats, their abilities are very much limited by the amount of context they receive. This [...] could cause [difficulties] for researchers who hope to use them to do things such as analyze ancient languages . In some cases, there are few historical records on long-gone civilizations to serve as training data for such a purpose." (p. 82.) Daniel Jurafsky and James H. Martin (2008). Speech and Language Processing , 2nd edition. Pearson Prentice Hall. ISBN 978-0-13-187321-6 . Mohamed Zakaria Kurdi (2016). Natural Language Processing and Computational Linguistics: speech, morphology, and syntax , Volume 1. ISTE-Wiley. ISBN 978-1848218482 . Mohamed Zakaria Kurdi (2017). Natural Language Processing and Computational Linguistics: semantics, discourse, and applications , Volume 2. ISTE-Wiley. ISBN 978-1848219212 . Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008). Introduction to Information Retrieval . Cambridge University Press. ISBN 978-0-521-86571-5 . Official html and pdf versions available without charge. Christopher D. Manning and Hinrich Schütze (1999). Foundations of Statistical Natural Language Processing . The MIT Press. ISBN 978-0-262-13360-9 . David M. W. Powers and Christopher C. R. Turk (1989). Machine Learning of Natural Language . Springer-Verlag. ISBN 978-0-387-19557-5 . Media related to Natural language processing at Wikimedia Commons
Markdown	[Jump to content](https://en.wikipedia.org/wiki/Natural_language_processing#bodyContent) Main menu Main menu move to sidebar hide Navigation - [Main page](https://en.wikipedia.org/wiki/Main_Page "Visit the main page [z]") - [Contents](https://en.wikipedia.org/wiki/Wikipedia:Contents "Guides to browsing Wikipedia") - [Current events](https://en.wikipedia.org/wiki/Portal:Current_events "Articles related to current events") - [Random article](https://en.wikipedia.org/wiki/Special:Random "Visit a randomly selected article [x]") - [About Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:About "Learn about Wikipedia and how it works") - [Contact us](https://en.wikipedia.org/wiki/Wikipedia:Contact_us "How to contact Wikipedia") Contribute - [Help](https://en.wikipedia.org/wiki/Help:Contents "Guidance on how to use and edit Wikipedia") - [Learn to edit](https://en.wikipedia.org/wiki/Help:Introduction "Learn how to edit Wikipedia") - [Community portal](https://en.wikipedia.org/wiki/Wikipedia:Community_portal "The hub for editors") - [Recent changes](https://en.wikipedia.org/wiki/Special:RecentChanges "A list of recent changes to Wikipedia [r]") - [Upload file](https://en.wikipedia.org/wiki/Wikipedia:File_upload_wizard "Add images or other media for use on Wikipedia") - [Special pages](https://en.wikipedia.org/wiki/Special:SpecialPages "A list of all special pages [q]") [![](https://en.wikipedia.org/static/images/icons/enwiki-25.svg) ![Wikipedia](https://en.wikipedia.org/static/images/mobile/copyright/wikipedia-wordmark-en-25.svg) ![The Free Encyclopedia](https://en.wikipedia.org/static/images/mobile/copyright/wikipedia-tagline-en-25.svg)](https://en.wikipedia.org/wiki/Main_Page) [Search](https://en.wikipedia.org/wiki/Special:Search "Search Wikipedia [f]") Appearance - [Donate](https://donate.wikimedia.org/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en) - [Create account](https://en.wikipedia.org/w/index.php?title=Special:CreateAccount&returnto=Natural+language+processing "You are encouraged to create an account and log in; however, it is not mandatory") - [Log in](https://en.wikipedia.org/w/index.php?title=Special:UserLogin&returnto=Natural+language+processing "You're encouraged to log in; however, it's not mandatory. [o]") Personal tools - [Donate](https://donate.wikimedia.org/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en) - [Create account](https://en.wikipedia.org/w/index.php?title=Special:CreateAccount&returnto=Natural+language+processing "You are encouraged to create an account and log in; however, it is not mandatory") - [Log in](https://en.wikipedia.org/w/index.php?title=Special:UserLogin&returnto=Natural+language+processing "You're encouraged to log in; however, it's not mandatory. [o]") ## Contents move to sidebar hide - [(Top)](https://en.wikipedia.org/wiki/Natural_language_processing) - [1 History](https://en.wikipedia.org/wiki/Natural_language_processing#History) Toggle History subsection - [1\.1 Symbolic NLP (1950s – early 1990s)](https://en.wikipedia.org/wiki/Natural_language_processing#Symbolic_NLP_\(1950s_%E2%80%93_early_1990s\)) - [1\.2 Statistical NLP (1990s–present)](https://en.wikipedia.org/wiki/Natural_language_processing#Statistical_NLP_\(1990s%E2%80%93present\)) - [2 Approaches: Symbolic, statistical, neural networks](https://en.wikipedia.org/wiki/Natural_language_processing#Approaches:_Symbolic,_statistical,_neural_networks) Toggle Approaches: Symbolic, statistical, neural networks subsection - [2\.1 Statistical approach](https://en.wikipedia.org/wiki/Natural_language_processing#Statistical_approach) - [2\.2 Neural networks](https://en.wikipedia.org/wiki/Natural_language_processing#Neural_networks) - [3 Common NLP tasks](https://en.wikipedia.org/wiki/Natural_language_processing#Common_NLP_tasks) Toggle Common NLP tasks subsection - [3\.1 Text and speech processing](https://en.wikipedia.org/wiki/Natural_language_processing#Text_and_speech_processing) - [3\.2 Morphological analysis](https://en.wikipedia.org/wiki/Natural_language_processing#Morphological_analysis) - [3\.3 Syntactic analysis](https://en.wikipedia.org/wiki/Natural_language_processing#Syntactic_analysis) - [3\.4 Lexical semantics (of individual words in context)](https://en.wikipedia.org/wiki/Natural_language_processing#Lexical_semantics_\(of_individual_words_in_context\)) - [3\.5 Relational semantics (semantics of individual sentences)](https://en.wikipedia.org/wiki/Natural_language_processing#Relational_semantics_\(semantics_of_individual_sentences\)) - [3\.6 Discourse (semantics beyond individual sentences)](https://en.wikipedia.org/wiki/Natural_language_processing#Discourse_\(semantics_beyond_individual_sentences\)) - [3\.7 Higher-level NLP applications](https://en.wikipedia.org/wiki/Natural_language_processing#Higher-level_NLP_applications) - [4 General tendencies and (possible) future directions](https://en.wikipedia.org/wiki/Natural_language_processing#General_tendencies_and_\(possible\)_future_directions) Toggle General tendencies and (possible) future directions subsection - [4\.1 Cognition](https://en.wikipedia.org/wiki/Natural_language_processing#Cognition) - [5 See also](https://en.wikipedia.org/wiki/Natural_language_processing#See_also) - [6 References](https://en.wikipedia.org/wiki/Natural_language_processing#References) - [7 Further reading](https://en.wikipedia.org/wiki/Natural_language_processing#Further_reading) - [8 External links](https://en.wikipedia.org/wiki/Natural_language_processing#External_links) Toggle the table of contents # Natural language processing 71 languages - [Afrikaans](https://af.wikipedia.org/wiki/Natuurliketaalverwerking "Natuurliketaalverwerking – Afrikaans") - [العربية](https://ar.wikipedia.org/wiki/%D9%85%D8%B9%D8%A7%D9%84%D8%AC%D8%A9_%D8%A7%D9%84%D9%84%D8%BA%D8%A9_%D8%A7%D9%84%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A%D8%A9 "معالجة اللغة الطبيعية – Arabic") - [مصرى](https://arz.wikipedia.org/wiki/%D8%AA%D8%AD%D9%84%D9%8A%D9%84_%D8%A7%D9%84%D9%84%D8%BA%D8%A7%D8%AA_%D8%A7%D9%84%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A%D9%87 "تحليل اللغات الطبيعيه – Egyptian Arabic") - [Azərbaycanca](https://az.wikipedia.org/wiki/T%C9%99bii_dilin_emal%C4%B1 "Təbii dilin emalı – Azerbaijani") - [Беларуская (тарашкевіца)](https://be-tarask.wikipedia.org/wiki/%D0%90%D0%BF%D1%80%D0%B0%D1%86%D0%BE%D1%9E%D0%BA%D0%B0_%D0%BD%D0%B0%D1%82%D1%83%D1%80%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D0%B9_%D0%BC%D0%BE%D0%B2%D1%8B "Апрацоўка натуральнай мовы – Belarusian (Taraškievica orthography)") - [Беларуская](https://be.wikipedia.org/wiki/%D0%90%D0%BF%D1%80%D0%B0%D1%86%D0%BE%D1%9E%D0%BA%D0%B0_%D0%BD%D0%B0%D1%82%D1%83%D1%80%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D0%B9_%D0%BC%D0%BE%D0%B2%D1%8B "Апрацоўка натуральнай мовы – Belarusian") - [Български](https://bg.wikipedia.org/wiki/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%BD%D0%B0_%D0%B5%D1%81%D1%82%D0%B5%D1%81%D1%82%D0%B2%D0%B5%D0%BD_%D0%B5%D0%B7%D0%B8%D0%BA "Обработка на естествен език – Bulgarian") - [বাংলা](https://bn.wikipedia.org/wiki/%E0%A6%B8%E0%A7%8D%E0%A6%AC%E0%A6%BE%E0%A6%AD%E0%A6%BE%E0%A6%AC%E0%A6%BF%E0%A6%95_%E0%A6%AD%E0%A6%BE%E0%A6%B7%E0%A6%BE_%E0%A6%AA%E0%A7%8D%E0%A6%B0%E0%A6%95%E0%A7%8D%E0%A6%B0%E0%A6%BF%E0%A6%AF%E0%A6%BC%E0%A6%BE%E0%A6%9C%E0%A6%BE%E0%A6%A4%E0%A6%95%E0%A6%B0%E0%A6%A3 "স্বাভাবিক ভাষা প্রক্রিয়াজাতকরণ – Bangla") - [Brezhoneg](https://br.wikipedia.org/wiki/Treterezh_emgefre_al_lavar "Treterezh emgefre al lavar – Breton") - [Bosanski](https://bs.wikipedia.org/wiki/Obrada_prirodnog_jezika "Obrada prirodnog jezika – Bosnian") - [Català](https://ca.wikipedia.org/wiki/Processament_del_llenguatge_natural "Processament del llenguatge natural – Catalan") - [کوردی](https://ckb.wikipedia.org/wiki/%D9%BE%DB%8E%D9%88%D8%A7%DA%98%DB%86%DA%A9%D8%B1%D8%AF%D9%86%DB%8C_%D8%B2%D9%85%D8%A7%D9%86%DB%8C_%D8%B3%D8%B1%D9%88%D8%B4%D8%AA%DB%8C "پێواژۆکردنی زمانی سروشتی – Central Kurdish") - [Čeština](https://cs.wikipedia.org/wiki/Zpracov%C3%A1n%C3%AD_p%C5%99irozen%C3%A9ho_jazyka "Zpracování přirozeného jazyka – Czech") - [Cymraeg](https://cy.wikipedia.org/wiki/Prosesu_Iaith_Naturiol "Prosesu Iaith Naturiol – Welsh") - [Dansk](https://da.wikipedia.org/wiki/Sprogteknologi "Sprogteknologi – Danish") - [Deutsch](https://de.wikipedia.org/wiki/Verarbeitung_nat%C3%BCrlicher_Sprache "Verarbeitung natürlicher Sprache – German") - [Ελληνικά](https://el.wikipedia.org/wiki/%CE%95%CF%80%CE%B5%CE%BE%CE%B5%CF%81%CE%B3%CE%B1%CF%83%CE%AF%CE%B1_%CF%86%CF%85%CF%83%CE%B9%CE%BA%CE%AE%CF%82_%CE%B3%CE%BB%CF%8E%CF%83%CF%83%CE%B1%CF%82 "Επεξεργασία φυσικής γλώσσας – Greek") - [Esperanto](https://eo.wikipedia.org/wiki/Natur-lingva_prilaborado "Natur-lingva prilaborado – Esperanto") - [Español](https://es.wikipedia.org/wiki/Procesamiento_de_lenguajes_naturales "Procesamiento de lenguajes naturales – Spanish") - [Eesti](https://et.wikipedia.org/wiki/Loomuliku_keele_t%C3%B6%C3%B6tlus "Loomuliku keele töötlus – Estonian") - [Euskara](https://eu.wikipedia.org/wiki/Hizkuntzaren_prozesamendu "Hizkuntzaren prozesamendu – Basque") - [فارسی](https://fa.wikipedia.org/wiki/%D9%BE%D8%B1%D8%AF%D8%A7%D8%B2%D8%B4_%D8%B2%D8%A8%D8%A7%D9%86%E2%80%8C%D9%87%D8%A7%DB%8C_%D8%B7%D8%A8%DB%8C%D8%B9%DB%8C "پردازش زبان‌های طبیعی – Persian") - [Suomi](https://fi.wikipedia.org/wiki/Luonnollisen_kielen_k%C3%A4sittely "Luonnollisen kielen käsittely – Finnish") - [Français](https://fr.wikipedia.org/wiki/Traitement_automatique_des_langues "Traitement automatique des langues – French") - [Gaeilge](https://ga.wikipedia.org/wiki/Pr%C3%B3ise%C3%A1il_teanga_n%C3%A1d%C3%BArtha "Próiseáil teanga nádúrtha – Irish") - [Galego](https://gl.wikipedia.org/wiki/Procesamento_da_linguaxe_natural "Procesamento da linguaxe natural – Galician") - [עברית](https://he.wikipedia.org/wiki/%D7%A2%D7%99%D7%91%D7%95%D7%93_%D7%A9%D7%A4%D7%94_%D7%98%D7%91%D7%A2%D7%99%D7%AA "עיבוד שפה טבעית – Hebrew") - [हिन्दी](https://hi.wikipedia.org/wiki/%E0%A4%AA%E0%A5%8D%E0%A4%B0%E0%A4%BE%E0%A4%95%E0%A5%83%E0%A4%A4%E0%A4%BF%E0%A4%95_%E0%A4%AD%E0%A4%BE%E0%A4%B7%E0%A4%BE_%E0%A4%B8%E0%A4%82%E0%A4%B8%E0%A4%BE%E0%A4%A7%E0%A4%A8 "प्राकृतिक भाषा संसाधन – Hindi") - [Hrvatski](https://hr.wikipedia.org/wiki/Obrada_prirodnog_jezika "Obrada prirodnog jezika – Croatian") - [Հայերեն](https://hy.wikipedia.org/wiki/%D4%B2%D5%B6%D5%A1%D5%AF%D5%A1%D5%B6_%D5%AC%D5%A5%D5%A6%D5%BE%D5%AB_%D5%B4%D5%B7%D5%A1%D5%AF%D5%B8%D6%82%D5%B4 "Բնական լեզվի մշակում – Armenian") - [Արեւմտահայերէն](https://hyw.wikipedia.org/wiki/%D4%B2%D5%B6%D5%A1%D5%AF%D5%A1%D5%B6_%D4%BC%D5%A5%D5%A6%D5%B8%D6%82%D5%AB_%D5%84%D5%B7%D5%A1%D5%AF%D5%B8%D6%82%D5%B4 "Բնական Լեզուի Մշակում – Western Armenian") - [Bahasa Indonesia](https://id.wikipedia.org/wiki/Pengolahan_bahasa_alami "Pengolahan bahasa alami – Indonesian") - [Ido](https://io.wikipedia.org/wiki/Procedo_pri_naturala_linguo "Procedo pri naturala linguo – Ido") - [Íslenska](https://is.wikipedia.org/wiki/M%C3%A1lgreining "Málgreining – Icelandic") - [Italiano](https://it.wikipedia.org/wiki/Elaborazione_del_linguaggio_naturale "Elaborazione del linguaggio naturale – Italian") - [日本語](https://ja.wikipedia.org/wiki/%E8%87%AA%E7%84%B6%E8%A8%80%E8%AA%9E%E5%87%A6%E7%90%86 "自然言語処理 – Japanese") - [ქართული](https://ka.wikipedia.org/wiki/%E1%83%91%E1%83%A3%E1%83%9C%E1%83%94%E1%83%91%E1%83%A0%E1%83%98%E1%83%95%E1%83%98_%E1%83%94%E1%83%9C%E1%83%98%E1%83%A1_%E1%83%93%E1%83%90%E1%83%9B%E1%83%A3%E1%83%A8%E1%83%90%E1%83%95%E1%83%94%E1%83%91%E1%83%90 "ბუნებრივი ენის დამუშავება – Georgian") - [Qaraqalpaqsha](https://kaa.wikipedia.org/wiki/T%C3%A1biyiy_tildi_qayta_islew "Tábiyiy tildi qayta islew – Kara-Kalpak") - [ಕನ್ನಡ](https://kn.wikipedia.org/wiki/%E0%B2%B8%E0%B2%B9%E0%B2%9C_%E0%B2%AD%E0%B2%BE%E0%B2%B7%E0%B2%BE_%E0%B2%B8%E0%B2%82%E0%B2%B8%E0%B3%8D%E0%B2%95%E0%B2%B0%E0%B2%A3%E0%B3%86 "ಸಹಜ ಭಾಷಾ ಸಂಸ್ಕರಣೆ – Kannada") - [한국어](https://ko.wikipedia.org/wiki/%EC%9E%90%EC%97%B0%EC%96%B4_%EC%B2%98%EB%A6%AC "자연어 처리 – Korean") - [Lietuvių](https://lt.wikipedia.org/wiki/Nat%C5%ABraliosios_kalbos_apdorojimas "Natūraliosios kalbos apdorojimas – Lithuanian") - [Latviešu](https://lv.wikipedia.org/wiki/Dabisk%C4%81s_valodas_apstr%C4%81de "Dabiskās valodas apstrāde – Latvian") - [Македонски](https://mk.wikipedia.org/wiki/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%BD%D0%B0_%D0%BF%D1%80%D0%B8%D1%80%D0%BE%D0%B4%D0%BD%D0%B8_%D1%98%D0%B0%D0%B7%D0%B8%D1%86%D0%B8 "Обработка на природни јазици – Macedonian") - [Монгол](https://mn.wikipedia.org/wiki/%D0%9A%D0%BE%D0%BC%D0%BF%D1%8C%D1%8E%D1%82%D0%B5%D1%80%D1%8B%D0%BD_%D1%85%D1%8D%D0%BB_%D1%88%D0%B8%D0%BD%D0%B6%D0%BB%D1%8D%D0%BB "Компьютерын хэл шинжлэл – Mongolian") - [मराठी](https://mr.wikipedia.org/wiki/%E0%A4%A8%E0%A5%88%E0%A4%B8%E0%A4%B0%E0%A5%8D%E0%A4%97%E0%A4%BF%E0%A4%95_%E0%A4%AD%E0%A4%BE%E0%A4%B7%E0%A4%BE_%E0%A4%AA%E0%A5%8D%E0%A4%B0%E0%A4%95%E0%A5%8D%E0%A4%B0%E0%A4%BF%E0%A4%AF%E0%A4%BE "नैसर्गिक भाषा प्रक्रिया – Marathi") - [မြန်မာဘာသာ](https://my.wikipedia.org/wiki/%E1%80%99%E1%80%AD%E1%80%81%E1%80%84%E1%80%BA%E1%80%98%E1%80%AC%E1%80%9E%E1%80%AC%E1%80%85%E1%80%80%E1%80%AC%E1%80%B8%E1%80%9E%E1%80%AF%E1%80%B6%E1%80%B8_%E1%80%80%E1%80%BD%E1%80%94%E1%80%BA%E1%80%95%E1%80%BB%E1%80%B0%E1%80%90%E1%80%AC%E1%80%85%E1%80%94%E1%80%85%E1%80%BA "မိခင်ဘာသာစကားသုံး ကွန်ပျူတာစနစ် – Burmese") - [Nederlands](https://nl.wikipedia.org/wiki/Taaltechnologie "Taaltechnologie – Dutch") - [Norsk bokmål](https://no.wikipedia.org/wiki/Spr%C3%A5kteknologi "Språkteknologi – Norwegian Bokmål") - [ଓଡ଼ିଆ](https://or.wikipedia.org/wiki/%E0%AC%A8%E0%AD%8D%E0%AD%9F%E0%AC%BE%E0%AC%9A%E0%AD%81%E0%AC%B0%E0%AC%BE%E0%AC%B2_%E0%AC%B2%E0%AC%BE%E0%AC%99%E0%AD%8D%E0%AC%97%E0%AD%81%E0%AC%8F%E0%AC%9C_%E0%AC%AA%E0%AD%8D%E0%AC%B0%E0%AD%8B%E0%AC%B8%E0%AD%87%E0%AC%B8%E0%AC%BF%E0%AC%82 "ନ୍ୟାଚୁରାଲ ଲାଙ୍ଗୁଏଜ ପ୍ରୋସେସିଂ – Odia") - [Picard](https://pcd.wikipedia.org/wiki/Traitemint_automatique_d%27ches_langues "Traitemint automatique d'ches langues – Picard") - [Polski](https://pl.wikipedia.org/wiki/Przetwarzanie_j%C4%99zyka_naturalnego "Przetwarzanie języka naturalnego – Polish") - [Piemontèis](https://pms.wikipedia.org/wiki/NLP "NLP – Piedmontese") - [پښتو](https://ps.wikipedia.org/wiki/%D8%AF_%D8%B7%D8%A8%D9%8A%D8%B9%D9%8A_%DA%98%D8%A8%DB%90_%D9%BE%D8%B1%D9%88%D8%B3%D8%B3_%DA%A9%D9%88%D9%84 "د طبيعي ژبې پروسس کول – Pashto") - [Português](https://pt.wikipedia.org/wiki/Processamento_de_linguagem_natural "Processamento de linguagem natural – Portuguese") - [Runa Simi](https://qu.wikipedia.org/wiki/Paqariq_simi_thatkichay "Paqariq simi thatkichay – Quechua") - [Română](https://ro.wikipedia.org/wiki/Prelucrarea_limbajului_natural "Prelucrarea limbajului natural – Romanian") - [Русский](https://ru.wikipedia.org/wiki/%D0%9E%D0%B1%D1%80%D0%B0%D0%B1%D0%BE%D1%82%D0%BA%D0%B0_%D0%B5%D1%81%D1%82%D0%B5%D1%81%D1%82%D0%B2%D0%B5%D0%BD%D0%BD%D0%BE%D0%B3%D0%BE_%D1%8F%D0%B7%D1%8B%D0%BA%D0%B0 "Обработка естественного языка – Russian") - [Srpskohrvatski / српскохрватски](https://sh.wikipedia.org/wiki/Obrada_prirodnih_jezika "Obrada prirodnih jezika – Serbo-Croatian") - [Simple English](https://simple.wikipedia.org/wiki/Natural_language_processing "Natural language processing – Simple English") - [Shqip](https://sq.wikipedia.org/wiki/P%C3%ABrpunimi_i_gjuh%C3%ABs_natyrore "Përpunimi i gjuhës natyrore – Albanian") - [Српски / srpski](https://sr.wikipedia.org/wiki/Obrada_prirodnih_jezika "Obrada prirodnih jezika – Serbian") - [தமிழ்](https://ta.wikipedia.org/wiki/%E0%AE%87%E0%AE%AF%E0%AE%B1%E0%AF%8D%E0%AE%95%E0%AF%88_%E0%AE%AE%E0%AF%8A%E0%AE%B4%E0%AE%BF_%E0%AE%AE%E0%AF%81%E0%AE%B1%E0%AF%88%E0%AE%AF%E0%AE%BE%E0%AE%95%E0%AF%8D%E0%AE%95%E0%AE%AE%E0%AF%8D "இயற்கை மொழி முறையாக்கம் – Tamil") - [తెలుగు](https://te.wikipedia.org/wiki/%E0%B0%B8%E0%B0%B9%E0%B0%9C_%E0%B0%AD%E0%B0%BE%E0%B0%B7%E0%B0%BE_%E0%B0%AA%E0%B1%8D%E0%B0%B0%E0%B0%95%E0%B1%8D%E0%B0%B0%E0%B0%BF%E0%B0%AF "సహజ భాషా ప్రక్రియ – Telugu") - [ไทย](https://th.wikipedia.org/wiki/%E0%B8%81%E0%B8%B2%E0%B8%A3%E0%B8%9B%E0%B8%A3%E0%B8%B0%E0%B8%A1%E0%B8%A7%E0%B8%A5%E0%B8%A0%E0%B8%B2%E0%B8%A9%E0%B8%B2%E0%B8%98%E0%B8%A3%E0%B8%A3%E0%B8%A1%E0%B8%8A%E0%B8%B2%E0%B8%95%E0%B8%B4 "การประมวลภาษาธรรมชาติ – Thai") - [Türkçe](https://tr.wikipedia.org/wiki/Do%C4%9Fal_dil_i%C5%9Fleme "Doğal dil işleme – Turkish") - [Українська](https://uk.wikipedia.org/wiki/%D0%9E%D0%B1%D1%80%D0%BE%D0%B1%D0%BA%D0%B0_%D0%BF%D1%80%D0%B8%D1%80%D0%BE%D0%B4%D0%BD%D0%BE%D1%97_%D0%BC%D0%BE%D0%B2%D0%B8 "Обробка природної мови – Ukrainian") - [Tiếng Việt](https://vi.wikipedia.org/wiki/X%E1%BB%AD_l%C3%BD_ng%C3%B4n_ng%E1%BB%AF_t%E1%BB%B1_nhi%C3%AAn "Xử lý ngôn ngữ tự nhiên – Vietnamese") - [閩南語 / Bân-lâm-gí](https://zh-min-nan.wikipedia.org/wiki/Ch%C5%AB-ji%C3%A2n_gi%C3%A2n-g%C3%BA_chh%C3%BA-l%C3%AD "Chū-jiân giân-gú chhú-lí – Minnan") - [粵語](https://zh-yue.wikipedia.org/wiki/%E8%87%AA%E7%84%B6%E8%AA%9E%E8%A8%80%E8%99%95%E7%90%86 "自然語言處理 – Cantonese") - [中文](https://zh.wikipedia.org/wiki/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86 "自然语言处理 – Chinese") - [IsiZulu](https://zu.wikipedia.org/wiki/Ukudludlunga_ulimi_lwemvelo "Ukudludlunga ulimi lwemvelo – Zulu") [Edit links](https://www.wikidata.org/wiki/Special:EntityPage/Q30642#sitelinks-wikipedia "Edit interlanguage links") - [Article](https://en.wikipedia.org/wiki/Natural_language_processing "View the content page [c]") - [Talk](https://en.wikipedia.org/wiki/Talk:Natural_language_processing "Discuss improvements to the content page [t]") English - [Read](https://en.wikipedia.org/wiki/Natural_language_processing) - [Edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit "Edit this page [e]") - [View history](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=history "Past revisions of this page [h]") Tools Tools move to sidebar hide Actions - [Read](https://en.wikipedia.org/wiki/Natural_language_processing) - [Edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit "Edit this page [e]") - [View history](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=history) General - [What links here](https://en.wikipedia.org/wiki/Special:WhatLinksHere/Natural_language_processing "List of all English Wikipedia pages containing links to this page [j]") - [Related changes](https://en.wikipedia.org/wiki/Special:RecentChangesLinked/Natural_language_processing "Recent changes in pages linked from this page [k]") - [Upload file](https://en.wikipedia.org/wiki/Wikipedia:File_Upload_Wizard "Upload files [u]") - [Permanent link](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&oldid=1343173705 "Permanent link to this revision of this page") - [Page information](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=info "More information about this page") - [Cite this page](https://en.wikipedia.org/w/index.php?title=Special:CiteThisPage&page=Natural_language_processing&id=1343173705&wpFormIdentifier=titleform "Information on how to cite this page") - [Get shortened URL](https://en.wikipedia.org/w/index.php?title=Special:UrlShortener&url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FNatural_language_processing) Print/export - [Download as PDF](https://en.wikipedia.org/w/index.php?title=Special:DownloadAsPdf&page=Natural_language_processing&action=show-download-screen "Download this page as a PDF file") - [Printable version](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&printable=yes "Printable version of this page [p]") In other projects - [Wikimedia Commons](https://commons.wikimedia.org/wiki/Category:Natural_language_processing) - [Wikiversity](https://en.wikiversity.org/wiki/Natural_Language_Processing) - [Wikidata item](https://www.wikidata.org/wiki/Special:EntityPage/Q30642 "Structured data on this page hosted by Wikidata [g]") Appearance move to sidebar hide From Wikipedia, the free encyclopedia Processing of natural language by a computer \| \| \| \|---\|---\| \| ![](https://upload.wikimedia.org/wikipedia/en/thumb/b/b4/Ambox_important.svg/40px-Ambox_important.svg.png) \| \| \| \| \| \| [![icon](https://upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/60px-Question_book-new.svg.png)](https://en.wikipedia.org/wiki/File:Question_book-new.svg) \| This article needs additional citations for [verification](https://en.wikipedia.org/wiki/Wikipedia:Verifiability "Wikipedia:Verifiability"). Please help [improve this article](https://en.wikipedia.org/wiki/Special:EditPage/Natural_language_processing "Special:EditPage/Natural language processing") by [adding citations to reliable sources](https://en.wikipedia.org/wiki/Help:Referencing_for_beginners "Help:Referencing for beginners"). Unsourced material may be challenged and removed. Find sources: ["Natural language processing"](https://www.google.com/search?as_eq=wikipedia&q=%22Natural+language+processing%22) – [news](https://www.google.com/search?tbm=nws&q=%22Natural+language+processing%22+-wikipedia&tbs=ar:1) · [newspapers](https://www.google.com/search?&q=%22Natural+language+processing%22&tbs=bkt:s&tbm=bks) · [books](https://www.google.com/search?tbs=bks:1&q=%22Natural+language+processing%22+-wikipedia) · [scholar](https://scholar.google.com/scholar?q=%22Natural+language+processing%22) · [JSTOR](https://www.jstor.org/action/doBasicSearch?Query=%22Natural+language+processing%22&acc=on&wc=on) (May 2024) ([Learn how and when to remove this message](https://en.wikipedia.org/wiki/Help:Maintenance_template_removal "Help:Maintenance template removal")) \| \| \| \| \| [![icon](https://upload.wikimedia.org/wikipedia/commons/thumb/e/e8/Crystal_Clear_app_kedit.svg/40px-Crystal_Clear_app_kedit.svg.png)](https://en.wikipedia.org/wiki/File:Crystal_Clear_app_kedit.svg) \| This article may need to be rewritten to comply with Wikipedia's [quality standards](https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style "Wikipedia:Manual of Style"). Relevant discussion may be found on the [talk page](https://en.wikipedia.org/wiki/Talk:Natural_language_processing "Talk:Natural language processing"). [You can help](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit). The talk page may contain suggestions. (July 2025) \| \| \| \| \| ![](https://upload.wikimedia.org/wikipedia/commons/thumb/1/1c/Ambox_rewrite.svg/40px-Ambox_rewrite.svg.png) \| This article may be in need of reorganization to comply with Wikipedia's [layout guidelines](https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style/Layout "Wikipedia:Manual of Style/Layout"). Please help by [editing the article](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit) to make improvements to the overall structure. (July 2025) ([Learn how and when to remove this message](https://en.wikipedia.org/wiki/Help:Maintenance_template_removal "Help:Maintenance template removal")) \| \| \| \|---\| \| Part of [a series](https://en.wikipedia.org/wiki/Category:Artificial_intelligence "Category:Artificial intelligence") on \| \| [Artificial intelligence (AI)](https://en.wikipedia.org/wiki/Artificial_intelligence "Artificial intelligence") \| \| [![](https://upload.wikimedia.org/wikipedia/commons/thumb/6/64/Dall-e_3_%28jan_%2724%29_artificial_intelligence_icon.png/120px-Dall-e_3_%28jan_%2724%29_artificial_intelligence_icon.png)](https://en.wikipedia.org/wiki/File:Dall-e_3_\(jan_%2724\)_artificial_intelligence_icon.png) \| \| [Major goals](https://en.wikipedia.org/wiki/Artificial_intelligence#Goals "Artificial intelligence") [Artificial general intelligence](https://en.wikipedia.org/wiki/Artificial_general_intelligence "Artificial general intelligence") [Intelligent agent](https://en.wikipedia.org/wiki/Intelligent_agent "Intelligent agent") [Recursive self-improvement](https://en.wikipedia.org/wiki/Recursive_self-improvement "Recursive self-improvement") [Planning](https://en.wikipedia.org/wiki/Automated_planning_and_scheduling "Automated planning and scheduling") [Computer vision](https://en.wikipedia.org/wiki/Computer_vision "Computer vision") [General game playing](https://en.wikipedia.org/wiki/General_game_playing "General game playing") [Knowledge representation](https://en.wikipedia.org/wiki/Knowledge_representation_and_reasoning "Knowledge representation and reasoning") [Natural language processing]() [Robotics](https://en.wikipedia.org/wiki/Robotics "Robotics") [AI safety](https://en.wikipedia.org/wiki/AI_safety "AI safety") \| \| Approaches [Machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning") [Symbolic](https://en.wikipedia.org/wiki/Symbolic_artificial_intelligence "Symbolic artificial intelligence") [Deep learning](https://en.wikipedia.org/wiki/Deep_learning "Deep learning") [Bayesian networks](https://en.wikipedia.org/wiki/Bayesian_network "Bayesian network") [Evolutionary algorithms](https://en.wikipedia.org/wiki/Evolutionary_algorithm "Evolutionary algorithm") [Hybrid intelligent systems](https://en.wikipedia.org/wiki/Hybrid_intelligent_system "Hybrid intelligent system") [Systems integration](https://en.wikipedia.org/wiki/Artificial_intelligence_systems_integration "Artificial intelligence systems integration") [Open-source](https://en.wikipedia.org/wiki/Open-source_artificial_intelligence "Open-source artificial intelligence") [AI data centers](https://en.wikipedia.org/wiki/AI_data_center "AI data center") \| \| [Applications](https://en.wikipedia.org/wiki/Applications_of_artificial_intelligence "Applications of artificial intelligence") [Bioinformatics](https://en.wikipedia.org/wiki/Machine_learning_in_bioinformatics "Machine learning in bioinformatics") [Deepfake](https://en.wikipedia.org/wiki/Deepfake "Deepfake") [Earth sciences](https://en.wikipedia.org/wiki/Machine_learning_in_earth_sciences "Machine learning in earth sciences") [Finance](https://en.wikipedia.org/wiki/Applications_of_artificial_intelligence#Finance "Applications of artificial intelligence") [Generative AI](https://en.wikipedia.org/wiki/Generative_artificial_intelligence "Generative artificial intelligence") [Art](https://en.wikipedia.org/wiki/Artificial_intelligence_art "Artificial intelligence art") [Audio](https://en.wikipedia.org/wiki/Generative_audio "Generative audio") [Music](https://en.wikipedia.org/wiki/Music_and_artificial_intelligence "Music and artificial intelligence") [Government](https://en.wikipedia.org/wiki/Artificial_intelligence_in_government "Artificial intelligence in government") [Healthcare](https://en.wikipedia.org/wiki/Artificial_intelligence_in_healthcare "Artificial intelligence in healthcare") [Mental health](https://en.wikipedia.org/wiki/Artificial_intelligence_in_mental_health "Artificial intelligence in mental health") [Industry](https://en.wikipedia.org/wiki/Artificial_intelligence_in_industry "Artificial intelligence in industry") [Software development](https://en.wikipedia.org/wiki/AI-assisted_software_development "AI-assisted software development") [Translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation") [Military](https://en.wikipedia.org/wiki/Artificial_intelligence_arms_race "Artificial intelligence arms race") [Physics](https://en.wikipedia.org/wiki/Machine_learning_in_physics "Machine learning in physics") [Projects](https://en.wikipedia.org/wiki/List_of_artificial_intelligence_projects "List of artificial intelligence projects") \| \| [Philosophy](https://en.wikipedia.org/wiki/Philosophy_of_artificial_intelligence "Philosophy of artificial intelligence") [AI alignment](https://en.wikipedia.org/wiki/AI_alignment "AI alignment") [Artificial consciousness](https://en.wikipedia.org/wiki/Artificial_consciousness "Artificial consciousness") [The bitter lesson](https://en.wikipedia.org/wiki/Bitter_lesson "Bitter lesson") [Chinese room](https://en.wikipedia.org/wiki/Chinese_room "Chinese room") [Friendly AI](https://en.wikipedia.org/wiki/Friendly_artificial_intelligence "Friendly artificial intelligence") [Ethics](https://en.wikipedia.org/wiki/Ethics_of_artificial_intelligence "Ethics of artificial intelligence") [Existential risk](https://en.wikipedia.org/wiki/Existential_risk_from_artificial_general_intelligence "Existential risk from artificial general intelligence") [Turing test](https://en.wikipedia.org/wiki/Turing_test "Turing test") [Uncanny valley](https://en.wikipedia.org/wiki/Uncanny_valley "Uncanny valley") [Human–AI interaction](https://en.wikipedia.org/wiki/Human%E2%80%93AI_interaction "Human–AI interaction") \| \| [History](https://en.wikipedia.org/wiki/History_of_artificial_intelligence "History of artificial intelligence") [Timeline](https://en.wikipedia.org/wiki/Timeline_of_artificial_intelligence "Timeline of artificial intelligence") [Progress](https://en.wikipedia.org/wiki/Progress_in_artificial_intelligence "Progress in artificial intelligence") [AI winter](https://en.wikipedia.org/wiki/AI_winter "AI winter") [AI boom](https://en.wikipedia.org/wiki/AI_boom "AI boom") [AI bubble](https://en.wikipedia.org/wiki/AI_bubble "AI bubble") \| \| [Controversies](https://en.wikipedia.org/wiki/Artificial_intelligence_controversies "Artificial intelligence controversies") [Deepfake pornography](https://en.wikipedia.org/wiki/Deepfake_pornography "Deepfake pornography") [Taylor Swift deepfake pornography controversy](https://en.wikipedia.org/wiki/Taylor_Swift_deepfake_pornography_controversy "Taylor Swift deepfake pornography controversy") [Grok sexual deepfake scandal](https://en.wikipedia.org/wiki/Grok_sexual_deepfake_scandal "Grok sexual deepfake scandal") [Google Gemini image generation controversy](https://en.wikipedia.org/wiki/Google_Gemini_image_generation_controversy "Google Gemini image generation controversy") [It's the Most Terrible Time of the Year](https://en.wikipedia.org/wiki/It%27s_the_Most_Terrible_Time_of_the_Year "It's the Most Terrible Time of the Year") [Pause Giant AI Experiments](https://en.wikipedia.org/wiki/Pause_Giant_AI_Experiments:_An_Open_Letter "Pause Giant AI Experiments: An Open Letter") [Removal of Sam Altman from OpenAI](https://en.wikipedia.org/wiki/Removal_of_Sam_Altman_from_OpenAI "Removal of Sam Altman from OpenAI") [Statement on AI Risk](https://en.wikipedia.org/wiki/Statement_on_AI_Risk "Statement on AI Risk") [Tay (chatbot)](https://en.wikipedia.org/wiki/Tay_\(chatbot\) "Tay (chatbot)") [Théâtre D'opéra Spatial](https://en.wikipedia.org/wiki/Th%C3%A9%C3%A2tre_D%27op%C3%A9ra_Spatial "Théâtre D'opéra Spatial") [Voiceverse NFT plagiarism scandal](https://en.wikipedia.org/wiki/Voiceverse_NFT_plagiarism_scandal "Voiceverse NFT plagiarism scandal") \| \| Glossary [Glossary](https://en.wikipedia.org/wiki/Glossary_of_artificial_intelligence "Glossary of artificial intelligence") \| \| [v](https://en.wikipedia.org/wiki/Template:Artificial_intelligence "Template:Artificial intelligence") [t](https://en.wikipedia.org/wiki/Template_talk:Artificial_intelligence "Template talk:Artificial intelligence") [e](https://en.wikipedia.org/wiki/Special:EditPage/Template:Artificial_intelligence "Special:EditPage/Template:Artificial intelligence") \| Natural language processing (NLP) is the processing of [natural language](https://en.wikipedia.org/wiki/Natural_language "Natural language") information by a [computer](https://en.wikipedia.org/wiki/Computer "Computer"). NLP is a subfield of [computer science](https://en.wikipedia.org/wiki/Computer_science "Computer science") and is closely associated with [artificial intelligence](https://en.wikipedia.org/wiki/Artificial_intelligence "Artificial intelligence"). NLP is also related to [information retrieval](https://en.wikipedia.org/wiki/Information_retrieval "Information retrieval"), [knowledge representation](https://en.wikipedia.org/wiki/Knowledge_representation_and_reasoning "Knowledge representation and reasoning"), [computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics"), and [linguistics](https://en.wikipedia.org/wiki/Linguistics "Linguistics") more broadly.[\[1\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-nlpintro-1) Major processing tasks in an NLP system include: [speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition"), [text classification](https://en.wikipedia.org/wiki/Text_classification "Text classification"), [natural language understanding](https://en.wikipedia.org/wiki/Natural_language_understanding "Natural language understanding"), and [natural language generation](https://en.wikipedia.org/wiki/Natural_language_generation "Natural language generation"). ## History \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=1 "Edit section: History")\] Further information: [History of natural language processing](https://en.wikipedia.org/wiki/History_of_natural_language_processing "History of natural language processing") Natural language processing has its roots in the 1950s.[\[2\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-2) Already in 1950, [Alan Turing](https://en.wikipedia.org/wiki/Alan_Turing "Alan Turing") published an article titled "[Computing Machinery and Intelligence](https://en.wikipedia.org/wiki/Computing_Machinery_and_Intelligence "Computing Machinery and Intelligence")" which proposed what is now called the [Turing test](https://en.wikipedia.org/wiki/Turing_test "Turing test") as a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language. ### Symbolic NLP (1950s – early 1990s) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=2 "Edit section: Symbolic NLP (1950s – early 1990s)")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/0/09/AST_Document.svg/250px-AST_Document.svg.png)](https://en.wikipedia.org/wiki/File:AST_Document.svg) A document parsed into an abstract syntax tree The premise of symbolic NLP is often illustrated using [John Searle's Chinese room](https://en.wikipedia.org/wiki/John_Searle_\(American_philosopher\) "John Searle (American philosopher)") thought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts. - 1950s: The [Georgetown experiment](https://en.wikipedia.org/wiki/Georgetown-IBM_experiment "Georgetown-IBM experiment") in 1954 involved fully [automatic translation](https://en.wikipedia.org/wiki/Automatic_translation "Automatic translation") of more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem.[\[3\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-3) However, real progress was much slower, and after the [ALPAC report](https://en.wikipedia.org/wiki/ALPAC "ALPAC") in 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe[\[4\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-4)) until the late 1980s when the first [statistical machine translation](https://en.wikipedia.org/wiki/Statistical_machine_translation "Statistical machine translation") systems were developed. - 1960s: Some notably successful natural language processing systems developed in the 1960s were [SHRDLU](https://en.wikipedia.org/wiki/SHRDLU "SHRDLU"), a natural language system working in restricted "[blocks worlds](https://en.wikipedia.org/wiki/Blocks_world "Blocks world")" with restricted vocabularies, and [ELIZA](https://en.wikipedia.org/wiki/ELIZA "ELIZA"), a simulation of a [Rogerian psychotherapy](https://en.wikipedia.org/wiki/Rogerian_psychotherapy "Rogerian psychotherapy"), written by [Joseph Weizenbaum](https://en.wikipedia.org/wiki/Joseph_Weizenbaum "Joseph Weizenbaum") between 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the "patient" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to "My head hurts" with "Why do you say your head hurts?". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only twenty words, because that was all that would fit in a computer memory at the time.[\[5\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-5) - 1970s: During the 1970s, many programmers began to write "conceptual [ontologies](https://en.wikipedia.org/wiki/Ontology_\(information_science\) "Ontology (information science)")", which structured real-world information into computer-understandable data. Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981). During this time, the first [chatterbots](https://en.wikipedia.org/wiki/Chatterbots "Chatterbots") were written (e.g., [PARRY](https://en.wikipedia.org/wiki/PARRY "PARRY")). - 1980s: The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of [HPSG](https://en.wikipedia.org/wiki/Head-driven_phrase_structure_grammar "Head-driven phrase structure grammar") as a computational operationalization of [generative grammar](https://en.wikipedia.org/wiki/Generative_grammar "Generative grammar")), morphology (e.g., two-level morphology[\[6\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-6)), semantics (e.g., [Lesk algorithm](https://en.wikipedia.org/wiki/Lesk_algorithm "Lesk algorithm")), reference (e.g., within Centering Theory[\[7\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-7)) and other areas of natural language understanding (e.g., in the [Rhetorical Structure Theory](https://en.wikipedia.org/wiki/Rhetorical_structure_theory "Rhetorical structure theory")). Other lines of research were continued, e.g., the development of chatterbots with [Racter](https://en.wikipedia.org/wiki/Racter "Racter") and [Jabberwacky](https://en.wikipedia.org/wiki/Jabberwacky "Jabberwacky"). An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period.[\[8\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-8) ### Statistical NLP (1990s–present) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=3 "Edit section: Statistical NLP (1990s–present)")\] Up until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of [machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning") algorithms for language processing. This shift was influenced by increasing computational power (see [Moore's law](https://en.wikipedia.org/wiki/Moore%27s_law "Moore's law")) and a decline in the dominance of [Chomskyan](https://en.wikipedia.org/wiki/Chomskyan "Chomskyan") linguistic theories... (e.g. [transformational grammar](https://en.wikipedia.org/wiki/Transformational_grammar "Transformational grammar")), whose theoretical underpinnings discouraged the sort of [corpus linguistics](https://en.wikipedia.org/wiki/Corpus_linguistics "Corpus linguistics") that underlies the machine-learning approach to language processing.[\[9\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-9) - 1990s: Many of the notable early successes in statistical methods in NLP occurred in the field of [machine translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation"), due especially to work at IBM Research, such as [IBM alignment models](https://en.wikipedia.org/wiki/IBM_alignment_models "IBM alignment models"). These systems were able to take advantage of existing multilingual [textual corpora](https://en.wikipedia.org/wiki/Text_corpus "Text corpus") that had been produced by the [Parliament of Canada](https://en.wikipedia.org/wiki/Parliament_of_Canada "Parliament of Canada") and the [European Union](https://en.wikipedia.org/wiki/European_Union "European Union") as a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data. - 2000s: With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on [unsupervised](https://en.wikipedia.org/wiki/Unsupervised_learning "Unsupervised learning") and [semi-supervised learning](https://en.wikipedia.org/wiki/Semi-supervised_learning "Semi-supervised learning") algorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than [supervised learning](https://en.wikipedia.org/wiki/Supervised_learning "Supervised learning"), and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the [World Wide Web](https://en.wikipedia.org/wiki/World_Wide_Web "World Wide Web")), which can often make up for the worse efficiency if the algorithm used has a low enough [time complexity](https://en.wikipedia.org/wiki/Time_complexity "Time complexity") to be practical. - 2003: [word n-gram model](https://en.wikipedia.org/wiki/Word_n-gram_language_model "Word n-gram language model"), at the time the best statistical algorithm, is outperformed by a [multi-layer perceptron](https://en.wikipedia.org/wiki/Multi-layer_perceptron "Multi-layer perceptron") (with a single hidden layer and context length of several words, trained on up to 14 million words, by [Bengio](https://en.wikipedia.org/wiki/Yoshua_Bengio "Yoshua Bengio") et al.)[\[10\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-10) - 2010: [Tomáš Mikolov](https://en.wikipedia.org/wiki/Tom%C3%A1%C5%A1_Mikolov "Tomáš Mikolov") (then a PhD student at [Brno University of Technology](https://en.wikipedia.org/wiki/Brno_University_of_Technology "Brno University of Technology")) with co-authors applied a simple [recurrent neural network](https://en.wikipedia.org/wiki/Recurrent_neural_network "Recurrent neural network") with a single hidden layer to language modeling,[\[11\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-11) and in the following years he went on to develop [Word2vec](https://en.wikipedia.org/wiki/Word2vec "Word2vec"). In the 2010s, [representation learning](https://en.wikipedia.org/wiki/Representation_learning "Representation learning") and [deep neural network](https://en.wikipedia.org/wiki/Deep_learning "Deep learning")\-style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques[\[12\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-goldberg:nnlp17-12)[\[13\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-goodfellow:book16-13) can achieve state-of-the-art results in many natural language tasks, e.g., in [language modeling](https://en.wikipedia.org/wiki/Language_modeling "Language modeling")[\[14\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-jozefowicz:lm16-14) and parsing.[\[15\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-choe:emnlp16-15)[\[16\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-vinyals:nips15-16) This is increasingly important [in medicine and healthcare](https://en.wikipedia.org/wiki/Artificial_intelligence_in_healthcare "Artificial intelligence in healthcare"), where NLP helps analyze notes and text in [electronic health records](https://en.wikipedia.org/wiki/Electronic_health_record "Electronic health record") that would otherwise be inaccessible for study when seeking to improve care[\[17\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-17) or protect patient privacy.[\[18\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-18) ## Approaches: Symbolic, statistical, neural networks \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=4 "Edit section: Approaches: Symbolic, statistical, neural networks")\] Symbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular:[\[19\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-winograd:shrdlu71-19)[\[20\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-schank77-20) such as by writing grammars or devising heuristic rules for [stemming](https://en.wikipedia.org/wiki/Stemming "Stemming"). [Machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning") approaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach: - both statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally. - [language models](https://en.wikipedia.org/wiki/Language_model "Language model"), produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce. - the larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to [intractability](https://en.wikipedia.org/wiki/Intractable_problem "Intractable problem") problems. Rule-based systems are commonly used: - when the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the [Apertium](https://en.wikipedia.org/wiki/Apertium "Apertium") system, - for preprocessing in NLP pipelines, e.g., [tokenization](https://en.wikipedia.org/wiki/Tokenization_\(lexical_analysis\) "Tokenization (lexical analysis)"), or - for post-processing and transforming the output of NLP pipelines, e.g., for [knowledge extraction](https://en.wikipedia.org/wiki/Knowledge_extraction "Knowledge extraction") from syntactic parses. ### Statistical approach \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=5 "Edit section: Statistical approach")\] In the late 1980s and mid-1990s, the statistical approach ended a period of [AI winter](https://en.wikipedia.org/wiki/AI_winter "AI winter"), which was caused by the inefficiencies of the rule-based approaches.[\[21\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-johnson:eacl:ilcl09-21)[\[22\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-resnik:langlog11-22) The earliest [decision trees](https://en.wikipedia.org/wiki/Decision_tree "Decision tree"), producing systems of hard [if–then rules](https://en.wikipedia.org/wiki/Conditional_\(computer_programming\)#If%E2%80%93then\(%E2%80%93else\) "Conditional (computer programming)"), were still very similar to the old rule-based approaches. Only the introduction of hidden [Markov models](https://en.wikipedia.org/wiki/Markov_model "Markov model"), applied to part-of-speech tagging, announced the end of the old rule-based approach. ### Neural networks \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=6 "Edit section: Neural networks")\] Further information: [Artificial neural network](https://en.wikipedia.org/wiki/Artificial_neural_network "Artificial neural network") A major drawback of statistical methods is that they require elaborate [feature engineering](https://en.wikipedia.org/wiki/Feature_engineering "Feature engineering"). Since 2015,[\[23\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-23) [neural network](https://en.wikipedia.org/wiki/Neural_network_\(machine_learning\) "Neural network (machine learning)")–based methods have increasingly replaced traditional statistical approaches, using [semantic networks](https://en.wikipedia.org/wiki/Semantic_networks "Semantic networks")[\[24\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-24) and [word embeddings](https://en.wikipedia.org/wiki/Word_embedding "Word embedding") to capture semantic properties of words. Intermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore. [Neural machine translation](https://en.wikipedia.org/wiki/Neural_machine_translation "Neural machine translation"), based on then-newly invented [sequence-to-sequence](https://en.wikipedia.org/wiki/Seq2seq "Seq2seq") transformations, made obsolete the intermediate steps, such as word alignment, previously necessary for [statistical machine translation](https://en.wikipedia.org/wiki/Statistical_machine_translation "Statistical machine translation"). ## Common NLP tasks \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=7 "Edit section: Common NLP tasks")\] The following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks. Though natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below. ### Text and speech processing \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=8 "Edit section: Text and speech processing")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/8/8c/Hebrew_stop_words_word_cloud.png/250px-Hebrew_stop_words_word_cloud.png)](https://en.wikipedia.org/wiki/File:Hebrew_stop_words_word_cloud.png) Word cloud of stop words in Hebrew [Optical character recognition](https://en.wikipedia.org/wiki/Optical_character_recognition "Optical character recognition") (OCR) Given an image representing printed text, determine the corresponding text. [Speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition") Given a sound clip of a person or people speaking, determine the textual representation of the speech. This is the opposite of [text to speech](https://en.wikipedia.org/wiki/Text_to_speech "Text to speech") and is one of the extremely difficult problems colloquially termed "[AI-complete](https://en.wikipedia.org/wiki/AI-complete "AI-complete")" (see above). In [natural speech](https://en.wikipedia.org/wiki/Natural_speech "Natural speech") there are hardly any pauses between successive words, and thus [speech segmentation](https://en.wikipedia.org/wiki/Speech_segmentation "Speech segmentation") is a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed [coarticulation](https://en.wikipedia.org/wiki/Coarticulation "Coarticulation"), so the conversion of the [analog signal](https://en.wikipedia.org/wiki/Analog_signal "Analog signal") to discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent. [Speech segmentation](https://en.wikipedia.org/wiki/Speech_segmentation "Speech segmentation") Given a sound clip of a person or people speaking, separate it into words. A subtask of [speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition") and typically grouped with it. [Text-to-speech](https://en.wikipedia.org/wiki/Text-to-speech "Text-to-speech") Given a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired.[\[25\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-25) [Word segmentation](https://en.wikipedia.org/wiki/Word_segmentation "Word segmentation") ([Tokenization](https://en.wikipedia.org/wiki/Tokenization_\(lexical_analysis\) "Tokenization (lexical analysis)")) [Tokenization](https://en.wikipedia.org/wiki/Tokenization_\(data_security\) "Tokenization (data security)") is a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods.[\[26\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-:0-26) For a language like [English](https://en.wikipedia.org/wiki/English_language "English language"), this is fairly trivial, since words are usually separated by spaces. However, some written languages like [Chinese](https://en.wikipedia.org/wiki/Chinese_language "Chinese language"), [Japanese](https://en.wikipedia.org/wiki/Japanese_language "Japanese language") and [Thai](https://en.wikipedia.org/wiki/Thai_language "Thai language") do not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the [vocabulary](https://en.wikipedia.org/wiki/Vocabulary "Vocabulary") and [morphology](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") of words in the language. Sometimes this process is also used in cases like [bag of words](https://en.wikipedia.org/wiki/Bag_of_words "Bag of words") (BOW) creation in data mining.[\[27\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-27)[\[28\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-28) ### Morphological analysis \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=9 "Edit section: Morphological analysis")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/Eustagger_euskal_lematizatzailearen_adibidea.png/250px-Eustagger_euskal_lematizatzailearen_adibidea.png)](https://en.wikipedia.org/wiki/File:Eustagger_euskal_lematizatzailearen_adibidea.png) Lemmatization of Basque words [Lemmatization](https://en.wikipedia.org/wiki/Lemmatisation "Lemmatisation") The task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form.[\[29\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-29) [Morphological segmentation](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") Separate words into individual [morphemes](https://en.wikipedia.org/wiki/Morpheme "Morpheme") and identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the [morphology](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") (i.e., the structure of words) of the language being considered. [English](https://en.wikipedia.org/wiki/English_language "English language") has fairly simple morphology, especially [inflectional morphology](https://en.wikipedia.org/wiki/Inflectional_morphology "Inflectional morphology"), and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., "open, opens, opened, opening") as separate words. In languages such as [Turkish](https://en.wikipedia.org/wiki/Turkish_language "Turkish language") or [Meitei](https://en.wikipedia.org/wiki/Meitei_language "Meitei language"), a highly [agglutinated](https://en.wikipedia.org/wiki/Agglutination "Agglutination") Indian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms.[\[30\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-30) [Part-of-speech tagging](https://en.wikipedia.org/wiki/Part-of-speech_tagging "Part-of-speech tagging") Given a sentence, determine the [part of speech](https://en.wikipedia.org/wiki/Part_of_speech "Part of speech") (POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, "book" can be a [noun](https://en.wikipedia.org/wiki/Noun "Noun") ("the book on the table") or [verb](https://en.wikipedia.org/wiki/Verb "Verb") ("to book a flight"); "set" can be a noun, verb or [adjective](https://en.wikipedia.org/wiki/Adjective "Adjective"); and "out" can be any of at least five different parts of speech. [Stemming](https://en.wikipedia.org/wiki/Stemming "Stemming") The process of reducing inflected (or sometimes derived) words to a base form (e.g., "close" will be the root for "closed", "closing", "close", "closer" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary. ### Syntactic analysis \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=10 "Edit section: Syntactic analysis")\] \| \| \|---\| \| Part of [a series](https://en.wikipedia.org/wiki/Category:Formal_languages "Category:Formal languages") on \| \| [Formal languages](https://en.wikipedia.org/wiki/Formal_languages "Formal languages") \| \| Key concepts [Formal system](https://en.wikipedia.org/wiki/Formal_system "Formal system") [Alphabet](https://en.wikipedia.org/wiki/Alphabet_\(formal_languages\) "Alphabet (formal languages)") [Syntax](https://en.wikipedia.org/wiki/Syntax_\(logic\) "Syntax (logic)") [Formal semantics](https://en.wikipedia.org/wiki/Semantics_of_logic "Semantics of logic") [Semantics (programming languages)](https://en.wikipedia.org/wiki/Semantics_\(computer_science\) "Semantics (computer science)") [Formal grammar](https://en.wikipedia.org/wiki/Formal_grammar "Formal grammar") [Formation rule](https://en.wikipedia.org/wiki/Formation_rule "Formation rule") [Well-formed formula](https://en.wikipedia.org/wiki/Well-formed_formula "Well-formed formula") [Automata theory](https://en.wikipedia.org/wiki/Automata_theory "Automata theory") [Regular expression](https://en.wikipedia.org/wiki/Regular_expression "Regular expression") [Production](https://en.wikipedia.org/wiki/Production_\(computer_science\) "Production (computer science)") [Ground expression](https://en.wikipedia.org/wiki/Ground_expression "Ground expression") [Atomic formula](https://en.wikipedia.org/wiki/Atomic_formula "Atomic formula") \| \| Applications [Formal methods](https://en.wikipedia.org/wiki/Formal_methods "Formal methods") [Propositional calculus](https://en.wikipedia.org/wiki/Propositional_calculus "Propositional calculus") [Predicate logic](https://en.wikipedia.org/wiki/Predicate_logic "Predicate logic") [Mathematical notation](https://en.wikipedia.org/wiki/Mathematical_notation "Mathematical notation") [Natural language processing]() [Programming language theory](https://en.wikipedia.org/wiki/Programming_language_theory "Programming language theory") [Mathematical linguistics](https://en.wikipedia.org/wiki/Mathematical_linguistics "Mathematical linguistics") [Computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics") [Syntax analysis](https://en.wikipedia.org/wiki/Syntax_analysis "Syntax analysis") [Formal verification](https://en.wikipedia.org/wiki/Formal_verification "Formal verification") [Automated theorem proving](https://en.wikipedia.org/wiki/Automated_theorem_proving "Automated theorem proving") \| \| [v](https://en.wikipedia.org/wiki/Template:Formal_languages "Template:Formal languages") [t](https://en.wikipedia.org/w/index.php?title=Template_talk:Formal_languages&action=edit&redlink=1 "Template talk:Formal languages (page does not exist)") [e](https://en.wikipedia.org/wiki/Special:EditPage/Template:Formal_languages "Special:EditPage/Template:Formal languages") \| [Grammar induction](https://en.wikipedia.org/wiki/Grammar_induction "Grammar induction")[\[31\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-31) Generate a [formal grammar](https://en.wikipedia.org/wiki/Formal_grammar "Formal grammar") that describes a language's syntax. [Sentence breaking](https://en.wikipedia.org/wiki/Sentence_breaking "Sentence breaking") (also known as "[sentence boundary disambiguation](https://en.wikipedia.org/wiki/Sentence_boundary_disambiguation "Sentence boundary disambiguation")") Given a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by [periods](https://en.wikipedia.org/wiki/Full_stop "Full stop") or other [punctuation marks](https://en.wikipedia.org/wiki/Punctuation_mark "Punctuation mark"), but these same characters can serve other purposes (e.g., marking [abbreviations](https://en.wikipedia.org/wiki/Abbreviation "Abbreviation")). [Parsing](https://en.wikipedia.org/wiki/Parsing "Parsing") Determine the [parse tree](https://en.wikipedia.org/wiki/Parse_tree "Parse tree") (grammatical analysis) of a given sentence. The [grammar](https://en.wikipedia.org/wiki/Grammar "Grammar") for [natural languages](https://en.wikipedia.org/wiki/Natural_language "Natural language") is [ambiguous](https://en.wikipedia.org/wiki/Ambiguous "Ambiguous") and typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing: dependency parsing and constituency parsing. Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a [probabilistic context-free grammar](https://en.wikipedia.org/wiki/Probabilistic_context-free_grammar "Probabilistic context-free grammar") (PCFG) (see also [stochastic grammar](https://en.wikipedia.org/wiki/Stochastic_grammar "Stochastic grammar")). ### Lexical semantics (of individual words in context) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=11 "Edit section: Lexical semantics (of individual words in context)")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/3/34/Entity_Linking_-_Example_of_pipeline.png/250px-Entity_Linking_-_Example_of_pipeline.png)](https://en.wikipedia.org/wiki/File:Entity_Linking_-_Example_of_pipeline.png) An entity linking pipeline [Lexical semantics](https://en.wikipedia.org/wiki/Lexical_semantics "Lexical semantics") What is the computational meaning of individual words in context? [Distributional semantics](https://en.wikipedia.org/wiki/Distributional_semantics "Distributional semantics") How can we learn semantic representations from data? [Named entity recognition](https://en.wikipedia.org/wiki/Named_entity_recognition "Named entity recognition") (NER) Given a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although [capitalization](https://en.wikipedia.org/wiki/Capitalization "Capitalization") can aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of [named entity](https://en.wikipedia.org/wiki/Named_entity "Named entity"), and in any case, is often inaccurate or insufficient. For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized. Furthermore, many other languages in non-Western scripts (e.g. [Chinese](https://en.wikipedia.org/wiki/Chinese_language "Chinese language") or [Arabic](https://en.wikipedia.org/wiki/Arabic_language "Arabic language")) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example, [German](https://en.wikipedia.org/wiki/German_language "German language") capitalizes all [nouns](https://en.wikipedia.org/wiki/Noun "Noun"), regardless of whether they are names, and [French](https://en.wikipedia.org/wiki/French_language "French language") and [Spanish](https://en.wikipedia.org/wiki/Spanish_language "Spanish language") do not capitalize names that serve as [adjectives](https://en.wikipedia.org/wiki/Adjective "Adjective"). This task is also referred to as token classification.[\[32\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-32) [Sentiment analysis](https://en.wikipedia.org/wiki/Sentiment_analysis "Sentiment analysis") (see also [Multimodal sentiment analysis](https://en.wikipedia.org/wiki/Multimodal_sentiment_analysis "Multimodal sentiment analysis")) Sentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as [word n-grams](https://en.wikipedia.org/wiki/Word_n-gram_language_model "Word n-gram language model"), [Term Frequency-Inverse Document Frequency](https://en.wikipedia.org/wiki/Term_frequency-inverse_document_frequency "Term frequency-inverse document frequency") (TF-IDF) features, hand-generated features, or employ [deep learning](https://en.wikipedia.org/wiki/Deep_learning "Deep learning") models designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms.[\[26\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-:0-26) [Terminology extraction](https://en.wikipedia.org/wiki/Terminology_extraction "Terminology extraction") The goal of terminology extraction is to automatically extract relevant terms from a given corpus. [Word-sense disambiguation](https://en.wikipedia.org/wiki/Word-sense_disambiguation "Word-sense disambiguation") (WSD) Many words have more than one [meaning](https://en.wikipedia.org/wiki/Meaning_\(linguistics\) "Meaning (linguistics)"); we have to select the meaning which makes the most sense in context. For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as [WordNet](https://en.wikipedia.org/wiki/WordNet "WordNet"). [Entity linking](https://en.wikipedia.org/wiki/Entity_linking "Entity linking") Many words—typically proper names—refer to [named entities](https://en.wikipedia.org/wiki/Named_entity "Named entity"); here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context. ### Relational semantics (semantics of individual sentences) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=12 "Edit section: Relational semantics (semantics of individual sentences)")\] [Relationship extraction](https://en.wikipedia.org/wiki/Relationship_extraction "Relationship extraction") Given a chunk of text, identify the relationships among named entities (e.g. who is married to whom). [Semantic parsing](https://en.wikipedia.org/wiki/Semantic_parsing "Semantic parsing") Given a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in [AMR parsing](https://en.wikipedia.org/wiki/Abstract_Meaning_Representation "Abstract Meaning Representation")) or in accordance with a logical formalism (e.g., in [DRT parsing](https://en.wikipedia.org/wiki/Discourse_representation_theory "Discourse representation theory")). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see [Natural language understanding](https://en.wikipedia.org/wiki/Natural_language_processing#Natural_language_understanding) below). [Semantic role labelling](https://en.wikipedia.org/wiki/Semantic_role_labeling "Semantic role labeling") (see also implicit semantic role labelling below) Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https://en.wikipedia.org/wiki/Frame_semantics_\(linguistics\) "Frame semantics (linguistics)")), then identify and classify the frame elements ([semantic roles](https://en.wikipedia.org/wiki/Semantic_roles "Semantic roles")). ### Discourse (semantics beyond individual sentences) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=13 "Edit section: Discourse (semantics beyond individual sentences)")\] [Coreference resolution](https://en.wikipedia.org/wiki/Coreference "Coreference") Given a sentence or larger chunk of text, determine which words ("mentions") refer to the same objects ("entities"). [Anaphora resolution](https://en.wikipedia.org/wiki/Anaphora_resolution "Anaphora resolution") is a specific example of this task, and is specifically concerned with matching up [pronouns](https://en.wikipedia.org/wiki/Pronoun "Pronoun") with the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called "bridging relationships" involving [referring expressions](https://en.wikipedia.org/wiki/Referring_expression "Referring expression"). For example, in a sentence such as "He entered John's house through the front door", "the front door" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to). [Discourse analysis](https://en.wikipedia.org/wiki/Discourse_analysis "Discourse analysis") This rubric includes several related tasks. One task is discourse parsing, i.e., identifying the [discourse](https://en.wikipedia.org/wiki/Discourse "Discourse") structure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast). Another possible task is recognizing and classifying the [speech acts](https://en.wikipedia.org/wiki/Speech_act "Speech act") in a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.). Implicit semantic role labelling Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https://en.wikipedia.org/wiki/Frame_semantics_\(linguistics\) "Frame semantics (linguistics)")) and their explicit semantic roles in the current sentence (see [Semantic role labelling](https://en.wikipedia.org/wiki/Natural_language_processing#Semantic_role_labelling) above). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to [pro-drop languages](https://en.wikipedia.org/wiki/Pro-drop_language "Pro-drop language"). [Recognizing textual entailment](https://en.wikipedia.org/wiki/Textual_entailment "Textual entailment") Given two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false.[\[33\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-rte:11-33) [Topic segmentation](https://en.wikipedia.org/wiki/Topic_segmentation "Topic segmentation") and recognition Given a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment. [Argument mining](https://en.wikipedia.org/wiki/Argument_mining "Argument mining") The goal of argument mining is the automatic extraction and identification of argumentative structures from [natural language](https://en.wikipedia.org/wiki/Natural_language "Natural language") text with the aid of computer programs.[\[34\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-34) Such argumentative structures include the premise, conclusions, the [argument scheme](https://en.wikipedia.org/wiki/Argument_scheme "Argument scheme") and the relationship between the main and subsidiary argument, or the main and counter-argument within discourse.[\[35\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-35)[\[36\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-36) ### Higher-level NLP applications \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=14 "Edit section: Higher-level NLP applications")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/4/43/Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png/250px-Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)](https://en.wikipedia.org/wiki/File:Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png) Machine translation in Firefox [Automatic summarization](https://en.wikipedia.org/wiki/Automatic_summarization "Automatic summarization") (text summarization) Produce a readable summary of a chunk of text. Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper. Grammatical error correction Grammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011.[\[37\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-37)[\[38\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-38)[\[39\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-39) As far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as [GPT-2](https://en.wikipedia.org/wiki/GPT-2 "GPT-2"), this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications. [Logic translation](https://en.wikipedia.org/wiki/Logic_translation "Logic translation") Translate a text from a natural language into formal logic. [Machine translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation") (MT) Automatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed "[AI-complete](https://en.wikipedia.org/wiki/AI-complete "AI-complete")", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly. [Natural language understanding](https://en.wikipedia.org/wiki/Natural_language_understanding "Natural language understanding") (NLU) Convert chunks of text into more formal representations such as [first-order logic](https://en.wikipedia.org/wiki/First-order_logic "First-order logic") structures that are easier for [computer](https://en.wikipedia.org/wiki/Computer "Computer") programs to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as [closed-world assumption](https://en.wikipedia.org/wiki/Closed-world_assumption "Closed-world assumption") (CWA) vs. [open-world assumption](https://en.wikipedia.org/wiki/Open-world_assumption "Open-world assumption"), or subjective Yes/No vs. objective True/False is expected for the construction of a basis of semantics formalization.[\[40\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-40) [Natural language generation](https://en.wikipedia.org/wiki/Natural_language_generation "Natural language generation") (NLG): Convert information from computer databases or semantic intents into readable human language. Book generation Not an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter, The policeman's beard is half-constructed).[\[41\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-41) The first published work by a neural network was published in 2018, [1 the Road](https://en.wikipedia.org/wiki/1_the_Road "1 the Road"), marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free) [language models](https://en.wikipedia.org/wiki/Language_model "Language model"). The first machine-generated science book was published in 2019 (Beta Writer, Lithium-Ion Batteries, Springer, Cham).[\[42\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-42) Unlike Racter and 1 the Road, this is grounded on factual knowledge and based on text summarization. [Document AI](https://en.wikipedia.org/wiki/Document_AI "Document AI") A Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants.[\[43\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-43) [Dialogue management](https://en.wikipedia.org/wiki/Dialogue_system "Dialogue system") Computer systems intended to converse with a human. [Question answering](https://en.wikipedia.org/wiki/Question_answering "Question answering") Given a human-language question, determine its answer. Typical questions have a specific right answer (such as "What is the capital of Canada?"), but sometimes open-ended questions are also considered (such as "What is the meaning of life?"). [Text-to-image generation](https://en.wikipedia.org/wiki/Text-to-image_generation "Text-to-image generation") Given a description of an image, generate an image that matches the description.[\[44\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-44) Text-to-scene generation Given a description of a scene, generate a [3D model](https://en.wikipedia.org/wiki/3D_model "3D model") of the scene.[\[45\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-45)[\[46\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-46) [Text-to-video](https://en.wikipedia.org/wiki/Text-to-video_model "Text-to-video model") Given a description of a video, generate a video that matches the description.[\[47\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-47)[\[48\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-48) ## General tendencies and (possible) future directions \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=15 "Edit section: General tendencies and (possible) future directions")\] Based on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed:[\[49\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-49) - Interest on increasingly abstract, "cognitive" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09/2017–18: dependency syntax, 2004–05/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing). - Increasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+/100+ languages) - Elimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems) ### Cognition \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=16 "Edit section: Cognition")\] Most higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above). [Cognition](https://en.wikipedia.org/wiki/Cognition "Cognition") refers to "the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses."[\[50\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-50) [Cognitive science](https://en.wikipedia.org/wiki/Cognitive_science "Cognitive science") is the interdisciplinary, scientific study of the mind and its processes.[\[51\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-51) [Cognitive linguistics](https://en.wikipedia.org/wiki/Cognitive_linguistics "Cognitive linguistics") is an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics.[\[52\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-52) Especially during the age of [symbolic NLP](https://en.wikipedia.org/wiki/Natural_language_processing#Symbolic_NLP_\(1950s_%E2%80%93_early_1990s\)), the area of computational linguistics maintained strong ties with cognitive studies. As an example, [George Lakoff](https://en.wikipedia.org/wiki/George_Lakoff "George Lakoff") offers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics,[\[53\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-53) with two defining aspects: 1. Apply the theory of [conceptual metaphor](https://en.wikipedia.org/wiki/Conceptual_metaphor "Conceptual metaphor"), explained by Lakoff as "the understanding of one idea, in terms of another" which provides an idea of the intent of the author.[\[54\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-54) For example, consider the English word big. When used in a comparison ("That is a big tree"), the author's intent is to imply that the tree is physically large relative to other trees or the authors experience. When used metaphorically ("Tomorrow is a big day"), the author's intent to imply importance. The intent behind other usages, like in "She is a big person", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information. 2. Assign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a [probabilistic context-free grammar](https://en.wikipedia.org/wiki/Probabilistic_context-free_grammar "Probabilistic context-free grammar") (PCFG). The mathematical equation for such algorithms is presented in [US Patent 9269353](https://worldwide.espacenet.com/patent/search/family/055314712/publication/US9269353B1?q=pn%3DUS9269353):[\[55\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-55) R M M ( t o k e n N ) \= P M M ( t o k e n N ) × 1 2 d ( ∑ i \= − d d ( ( P M M ( t o k e n N ) × P F ( t o k e n N − i , t o k e n N , t o k e n N \+ i ) ) i ) {\\displaystyle {RMM(token\_{N})}={PMM(token\_{N})}\\times {\\frac {1}{2d}}\\left(\\sum \_{i=-d}^{d}{((PMM(token\_{N})}\\times {PF(token\_{N-i},token\_{N},token\_{N+i}))\_{i}}\\right)} ![{\\displaystyle {RMM(token\_{N})}={PMM(token\_{N})}\\times {\\frac {1}{2d}}\\left(\\sum \_{i=-d}^{d}{((PMM(token\_{N})}\\times {PF(token\_{N-i},token\_{N},token\_{N+i}))\_{i}}\\right)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/43ccadd794a4b84e20d1209997a463342e0dfbfe) Where RMM is the relative measure of meaning token is any block of text, sentence, phrase or word N is the number of tokens being analyzed PMM is the probable measure of meaning based on a corpora d is the non zero location of the token along the sequence of N tokens PF is the probability function specific to a language Ties with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar,[\[56\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-56) functional grammar,[\[57\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-57) construction grammar,[\[58\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-58) computational psycholinguistics and cognitive neuroscience (e.g., [ACT-R](https://en.wikipedia.org/wiki/ACT-R "ACT-R")), however, with limited uptake in mainstream NLP (as measured by presence on major conferences[\[59\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-59) of the [ACL](https://en.wikipedia.org/wiki/Association_for_Computational_Linguistics "Association for Computational Linguistics")). More recently, ideas of cognitive NLP have been revived as an approach to achieve [explainability](https://en.wikipedia.org/wiki/Explainable_artificial_intelligence "Explainable artificial intelligence"), e.g., under the notion of "cognitive AI".[\[60\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-60) Likewise, ideas of cognitive NLP are inherent to neural models [multimodal](https://en.wikipedia.org/wiki/Multimodal_interaction "Multimodal interaction") NLP (although rarely made explicit)[\[61\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-61) and developments in [artificial intelligence](https://en.wikipedia.org/wiki/Artificial_intelligence "Artificial intelligence"), specifically tools and technologies using [large language model](https://en.wikipedia.org/wiki/Large_language_model "Large language model") approaches[\[62\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-62) and new directions in [artificial general intelligence](https://en.wikipedia.org/wiki/Artificial_general_intelligence "Artificial general intelligence") based on the [free energy principle](https://en.wikipedia.org/wiki/Free_energy_principle "Free energy principle")[\[63\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-63) by British neuroscientist and theoretician at University College London [Karl J. Friston](https://en.wikipedia.org/wiki/Karl_J._Friston "Karl J. Friston"). ## See also \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=17 "Edit section: See also")\] - [1 the Road](https://en.wikipedia.org/wiki/1_the_Road "1 the Road") - [Artificial intelligence detection software](https://en.wikipedia.org/wiki/Artificial_intelligence_detection_software "Artificial intelligence detection software") - [Automated essay scoring](https://en.wikipedia.org/wiki/Automated_essay_scoring "Automated essay scoring") - [Biomedical text mining](https://en.wikipedia.org/wiki/Biomedical_text_mining "Biomedical text mining") - [Compound term processing](https://en.wikipedia.org/wiki/Compound_term_processing "Compound term processing") - [Computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics") - [Computer-assisted reviewing](https://en.wikipedia.org/wiki/Computer-assisted_reviewing "Computer-assisted reviewing") - [Controlled natural language](https://en.wikipedia.org/wiki/Controlled_natural_language "Controlled natural language") - [Deep learning](https://en.wikipedia.org/wiki/Deep_learning "Deep learning") - [Deep linguistic processing](https://en.wikipedia.org/wiki/Deep_linguistic_processing "Deep linguistic processing") - [Distributional semantics](https://en.wikipedia.org/wiki/Distributional_semantics "Distributional semantics") - [Foreign language reading aid](https://en.wikipedia.org/wiki/Foreign_language_reading_aid "Foreign language reading aid") - [Foreign language writing aid](https://en.wikipedia.org/wiki/Foreign_language_writing_aid "Foreign language writing aid") - [Information extraction](https://en.wikipedia.org/wiki/Information_extraction "Information extraction") - [Information retrieval](https://en.wikipedia.org/wiki/Information_retrieval "Information retrieval") - [Language and Communication Technologies](https://en.wikipedia.org/wiki/Language_and_Communication_Technologies "Language and Communication Technologies") - [Language model](https://en.wikipedia.org/wiki/Language_model "Language model") - [Language technology](https://en.wikipedia.org/wiki/Language_technology "Language technology") - [Latent semantic indexing](https://en.wikipedia.org/wiki/Latent_semantic_indexing "Latent semantic indexing") - [Multi-agent system](https://en.wikipedia.org/wiki/Multi-agent_system "Multi-agent system") - [Native-language identification](https://en.wikipedia.org/wiki/Native-language_identification "Native-language identification") - [Natural-language programming](https://en.wikipedia.org/wiki/Natural-language_programming "Natural-language programming") - [Natural-language understanding](https://en.wikipedia.org/wiki/Natural-language_understanding "Natural-language understanding") - [Natural-language search](https://en.wikipedia.org/wiki/Natural-language_user_interface "Natural-language user interface") - [Outline of natural language processing](https://en.wikipedia.org/wiki/Outline_of_natural_language_processing "Outline of natural language processing") - [Query expansion](https://en.wikipedia.org/wiki/Query_expansion "Query expansion") - [Query understanding](https://en.wikipedia.org/wiki/Query_understanding "Query understanding") - [Reification (linguistics)](https://en.wikipedia.org/wiki/Reification_\(linguistics\) "Reification (linguistics)") - [Speech processing](https://en.wikipedia.org/wiki/Speech_processing "Speech processing") - [Spoken dialogue systems](https://en.wikipedia.org/wiki/Spoken_dialogue_systems "Spoken dialogue systems") - [Text-proofing](https://en.wikipedia.org/wiki/Text-proofing "Text-proofing") - [Text simplification](https://en.wikipedia.org/wiki/Text_simplification "Text simplification") - [Transformer (machine learning model)](https://en.wikipedia.org/wiki/Transformer_\(machine_learning_model\) "Transformer (machine learning model)") - [Truecasing](https://en.wikipedia.org/wiki/Truecasing "Truecasing") - [Question answering](https://en.wikipedia.org/wiki/Question_answering "Question answering") - [Word2vec](https://en.wikipedia.org/wiki/Word2vec "Word2vec") ## References \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=18 "Edit section: References")\] 1. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-nlpintro_1-0) Eisenstein, Jacob (October 1, 2019). [Introduction to Natural Language Processing](https://mitpress.mit.edu/9780262042840/introduction-to-natural-language-processing/). The MIT Press. p. 1. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-04284-0](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-04284-0 "Special:BookSources/978-0-262-04284-0") . 2. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-2) ["NLP"](https://cs.stanford.edu/people/eroberts/courses/soco/projects/2004-05/nlp/overview_history.html). 3. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-3) Hutchins, J. (2005). ["The history of machine translation in a nutshell"](https://web.archive.org/web/20190713103044/http://www.hutchinsweb.me.uk/Nutshell-2005.pdf) (PDF). Archived from [the original](http://www.hutchinsweb.me.uk/Nutshell-2005.pdf) (PDF) on 2019-07-13. Retrieved 2019-02-04. \[[self-published source](https://en.wikipedia.org/wiki/Wikipedia:Verifiability#Self-published_sources "Wikipedia:Verifiability")\] 4. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-4) "ALPAC: the (in)famous report", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12. 5. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-5) [Crevier 1993](https://en.wikipedia.org/wiki/Natural_language_processing#CITEREFCrevier1993), pp. 146–148 harvnb error: no target: CITEREFCrevier1993 ([help](https://en.wikipedia.org/wiki/Category:Harv_and_Sfn_template_errors "Category:Harv and Sfn template errors")), see also [Buchanan 2005](https://en.wikipedia.org/wiki/Natural_language_processing#CITEREFBuchanan2005), p. 56 harvnb error: no target: CITEREFBuchanan2005 ([help](https://en.wikipedia.org/wiki/Category:Harv_and_Sfn_template_errors "Category:Harv and Sfn template errors")): "Early programs were necessarily limited in scope by the size and speed of memory" 6. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-6) [Koskenniemi, Kimmo](https://en.wikipedia.org/wiki/Kimmo_Koskenniemi "Kimmo Koskenniemi") (1983), [Two-level morphology: A general computational model of word-form recognition and production](https://web.archive.org/web/20181221032913/http://www.ling.helsinki.fi/~koskenni/doc/Two-LevelMorphology.pdf) (PDF), Department of General Linguistics, [University of Helsinki](https://en.wikipedia.org/wiki/University_of_Helsinki "University of Helsinki"), archived from [the original](http://www.ling.helsinki.fi/~koskenni/doc/Two-LevelMorphology.pdf) (PDF) on 2018-12-21, retrieved 2020-08-20 7. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-7) Joshi, A. K., & Weinstein, S. (1981, August). [Control of Inference: Role of Some Aspects of Discourse Structure-Centering](https://www.ijcai.org/Proceedings/81-1/Papers/071.pdf). In IJCAI (pp. 385–387). 8. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-8) Guida, G.; Mauri, G. (July 1986). "Evaluation of natural language processing systems: Issues and approaches". Proceedings of the IEEE. 74 (7): 1026–1035\. [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[1986IEEEP..74.1026G](https://ui.adsabs.harvard.edu/abs/1986IEEEP..74.1026G). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1109/PROC.1986.13580](https://doi.org/10.1109%2FPROC.1986.13580). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1558-2256](https://search.worldcat.org/issn/1558-2256). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [30688575](https://api.semanticscholar.org/CorpusID:30688575). 9. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-9) Chomskyan linguistics encourages the investigation of "[corner cases](https://en.wikipedia.org/wiki/Corner_case "Corner case")" that stress the limits of its theoretical models (comparable to [pathological](https://en.wikipedia.org/wiki/Pathological_\(mathematics\) "Pathological (mathematics)") phenomena in mathematics), typically created using [thought experiments](https://en.wikipedia.org/wiki/Thought_experiment "Thought experiment"), rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in [corpus linguistics](https://en.wikipedia.org/wiki/Corpus_linguistics "Corpus linguistics"). The creation and use of such [corpora](https://en.wikipedia.org/wiki/Text_corpus "Text corpus") of real-world data is a fundamental part of machine-learning algorithms for natural language processing. In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called "[poverty of the stimulus](https://en.wikipedia.org/wiki/Poverty_of_the_stimulus "Poverty of the stimulus")" argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing. As a result, the Chomskyan paradigm discouraged the application of such models to language processing. 10. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-10) Bengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003). ["A neural probabilistic language model"](https://dl.acm.org/doi/10.5555/944919.944966). The Journal of Machine Learning Research. 3: 1137–1155 – via ACM Digital Library. 11. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-11) Mikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010). ["Recurrent neural network based language model"](https://gwern.net/doc/ai/nn/rnn/2010-mikolov.pdf) (PDF). Interspeech 2010. pp. 1045–1048\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.21437/Interspeech.2010-343](https://doi.org/10.21437%2FInterspeech.2010-343). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [17048224](https://api.semanticscholar.org/CorpusID:17048224). `{{cite book}}`: `\|journal=` ignored ([help](https://en.wikipedia.org/wiki/Help:CS1_errors#periodical_ignored "Help:CS1 errors")) 12. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-goldberg:nnlp17_12-0) Goldberg, Yoav (2016). ["A Primer on Neural Network Models for Natural Language Processing"](https://doi.org/10.1613%2Fjair.4992). Journal of Artificial Intelligence Research. 57: 345–420\. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1807\.10854](https://arxiv.org/abs/1807.10854). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1613/jair.4992](https://doi.org/10.1613%2Fjair.4992). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [8273530](https://api.semanticscholar.org/CorpusID:8273530). 13. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-goodfellow:book16_13-0) Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016). [Deep Learning](http://www.deeplearningbook.org/). MIT Press. 14. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-jozefowicz:lm16_14-0) Jozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016). Exploring the Limits of Language Modeling. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1602\.02410](https://arxiv.org/abs/1602.02410). [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[2016arXiv160202410J](https://ui.adsabs.harvard.edu/abs/2016arXiv160202410J). 15. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-choe:emnlp16_15-0) Choe, Do Kook; Charniak, Eugene. ["Parsing as Language Modeling"](https://web.archive.org/web/20181023034804/https://aclanthology.coli.uni-saarland.de/papers/D16-1257/d16-1257). Emnlp 2016. Archived from [the original](https://aclanthology.coli.uni-saarland.de/papers/D16-1257/d16-1257) on 2018-10-23. Retrieved 2018-10-22. 16. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-vinyals:nips15_16-0) Vinyals, Oriol; et al. (2014). ["Grammar as a Foreign Language"](https://papers.nips.cc/paper/5635-grammar-as-a-foreign-language.pdf) (PDF). Nips2015. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1412\.7449](https://arxiv.org/abs/1412.7449). [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[2014arXiv1412.7449V](https://ui.adsabs.harvard.edu/abs/2014arXiv1412.7449V). 17. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-17) Turchin, Alexander; Florez Builes, Luisa F. (2021-03-19). ["Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120048). Journal of Diabetes Science and Technology. 15 (3): 553–560\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1177/19322968211000831](https://doi.org/10.1177%2F19322968211000831). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1932-2968](https://search.worldcat.org/issn/1932-2968). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [8120048](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120048). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [33736486](https://pubmed.ncbi.nlm.nih.gov/33736486). 18. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-18) Lee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10). ["Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9233261). JMIR Medical Informatics. 10 (6) e38482. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.2196/38482](https://doi.org/10.2196%2F38482). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [2291-9694](https://search.worldcat.org/issn/2291-9694). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [9233261](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9233261). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [35687381](https://pubmed.ncbi.nlm.nih.gov/35687381). 19. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-winograd:shrdlu71_19-0) Winograd, Terry (1971). [Procedures as a Representation for Data in a Computer Program for Understanding Natural Language](http://hci.stanford.edu/winograd/shrdlu/) (Thesis). 20. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-schank77_20-0) Schank, Roger C.; Abelson, Robert P. (1977). Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures. Hillsdale: Erlbaum. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [0-470-99033-3](https://en.wikipedia.org/wiki/Special:BookSources/0-470-99033-3 "Special:BookSources/0-470-99033-3") . 21. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-johnson:eacl:ilcl09_21-0) [Mark Johnson. How the statistical revolution changes (computational) linguistics.](http://www.aclweb.org/anthology/W09-0103) Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics. 22. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-resnik:langlog11_22-0) [Philip Resnik. Four revolutions.](http://languagelog.ldc.upenn.edu/nll/?p=2946) Language Log, February 5, 2011. 23. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-23) Socher, Richard. ["Deep Learning For NLP-ACL 2012 Tutorial"](https://web.archive.org/web/20210414054126/https://www.socher.org/index.php/Main/DeepLearningForNLP-ACL2012Tutorial). www.socher.org. Archived from [the original](https://www.socher.org/index.php/Main/DeepLearningForNLP-ACL2012Tutorial) on 2021-04-14. Retrieved 2020-08-17. This was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. \[Link is broken, try <http://web.stanford.edu/class/cs224n/>\] 24. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-24) Segev, Elad (2022). [Semantic Network Analysis in Social Sciences](https://www.routledge.com/Semantic-Network-Analysis-in-Social-Sciences/Segev/p/book/9780367636524). London: Routledge. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-367-63652-4](https://en.wikipedia.org/wiki/Special:BookSources/978-0-367-63652-4 "Special:BookSources/978-0-367-63652-4") . [Archived](https://web.archive.org/web/20211205140726/https://www.routledge.com/Semantic-Network-Analysis-in-Social-Sciences/Segev/p/book/9780367636524) from the original on 5 December 2021. Retrieved 5 December 2021. 25. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-25) Yi, Chucai; [Tian, Yingli](https://en.wikipedia.org/wiki/Yingli_Tian "Yingli Tian") (2012), "Assistive Text Reading from Complex Background for Blind Persons", Camera-Based Document Analysis and Recognition, Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. 15–28, [CiteSeerX](https://en.wikipedia.org/wiki/CiteSeerX_\(identifier\) "CiteSeerX (identifier)") [10\.1.1.668.869](https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.668.869), [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1007/978-3-642-29364-1\_2](https://doi.org/10.1007%2F978-3-642-29364-1_2), [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-3-642-29363-4](https://en.wikipedia.org/wiki/Special:BookSources/978-3-642-29363-4 "Special:BookSources/978-3-642-29363-4") `{{citation}}`: CS1 maint: work parameter with ISBN ([link](https://en.wikipedia.org/wiki/Category:CS1_maint:_work_parameter_with_ISBN "Category:CS1 maint: work parameter with ISBN")) 26. ^ [*a](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-:0_26-0) [b](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-:0_26-1) ["Natural Language Processing (NLP) - A Complete Guide"](https://www.deeplearning.ai/resources/natural-language-processing/). www.deeplearning.ai. 2023-01-11. Retrieved 2024-05-05. 27. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-27)* ["GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks"](https://www.geeksforgeeks.org/nlp/tokenization-in-natural-language-processing-nlp/). Geeksforgeeks. 25 June 2024. 28. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-28) Torabi, Mohhamadreza; Ghasemi, Fahimeh. ["Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12391535). Scientific Reports. 1 (1): 1–15\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1038/s41598-025-14699-1](https://doi.org/10.1038%2Fs41598-025-14699-1). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [12391535](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12391535). 29. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-29) ["What is Natural Language Processing? Intro to NLP in Machine Learning"](https://www.gyansetu.in/what-is-natural-language-processing/). GyanSetu\!. 2020-12-06. Retrieved 2021-01-09. 30. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-30) Kishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012). ["Manipuri Morpheme Identification"](http://aclweb.org/anthology//W/W12/W12-5008.pdf) (PDF). Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP). COLING 2012, Mumbai, December 2012: 95–108\. `{{cite journal}}`: CS1 maint: location ([link](https://en.wikipedia.org/wiki/Category:CS1_maint:_location "Category:CS1 maint: location")) 31. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-31) Klein, Dan; Manning, Christopher D. (2002). ["Natural language grammar induction using a constituent-context model"](http://papers.nips.cc/paper/1945-natural-language-grammar-induction-using-a-constituent-context-model.pdf) (PDF). Advances in Neural Information Processing Systems. 32. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-32) Kariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023). ["Precision information extraction for rare disease epidemiology at scale"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9972634). Journal of Translational Medicine. 21 (1): 157. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1186/s12967-023-04011-y](https://doi.org/10.1186%2Fs12967-023-04011-y). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [9972634](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9972634). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [36855134](https://pubmed.ncbi.nlm.nih.gov/36855134). 33. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-rte:11_33-0) PASCAL Recognizing Textual Entailment Challenge (RTE-7) <https://tac.nist.gov//2011/RTE/> 34. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-34) Lippi, Marco; Torroni, Paolo (2016-04-20). ["Argumentation Mining: State of the Art and Emerging Trends"](https://dl.acm.org/doi/10.1145/2850417). ACM Transactions on Internet Technology. 16 (2): 1–25\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1145/2850417](https://doi.org/10.1145%2F2850417). [hdl](https://en.wikipedia.org/wiki/Hdl_\(identifier\) "Hdl (identifier)"):[11585/523460](https://hdl.handle.net/11585%2F523460). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1533-5399](https://search.worldcat.org/issn/1533-5399). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [9561587](https://api.semanticscholar.org/CorpusID:9561587). 35. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-35) ["Argument Mining – IJCAI2016 Tutorial"](https://web.archive.org/web/20210418083659/https://www.i3s.unice.fr/~villata/tutorialIJCAI2016.html). www.i3s.unice.fr. Archived from [the original](https://www.i3s.unice.fr/~villata/tutorialIJCAI2016.html) on 2021-04-18. Retrieved 2021-03-09. 36. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-36) ["NLP Approaches to Computational Argumentation – ACL 2016, Berlin"](http://acl2016tutorial.arg.tech/). Retrieved 2021-03-09. 37. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-37) Administration. ["Centre for Language Technology (CLT)"](https://www.mq.edu.au/research/research-centres-groups-and-facilities/innovative-technologies/centres/centre-for-language-technology-clt). Macquarie University. Retrieved 2021-01-11. 38. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-38) ["Shared Task: Grammatical Error Correction"](https://www.comp.nus.edu.sg/~nlp/conll13st.html). www.comp.nus.edu.sg. Retrieved 2021-01-11. 39. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-39) ["Shared Task: Grammatical Error Correction"](https://www.comp.nus.edu.sg/~nlp/conll14st.html). www.comp.nus.edu.sg. Retrieved 2021-01-11. 40. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-40) Duan, Yucong; Cruz, Christophe (2011). ["Formalizing Semantic of Natural Language through Conceptualization from Existence"](https://web.archive.org/web/20111009135952/http://www.ijimt.org/abstract/100-E00187.htm). International Journal of Innovation, Management and Technology. 2 (1): 37–42\. Archived from [the original](http://www.ijimt.org/abstract/100-E00187.htm) on 2011-10-09. 41. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-41) ["U B U W E B :: Racter"](http://www.ubu.com/historical/racter/index.html). www.ubu.com. Retrieved 2020-08-17. 42. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-42) Writer, Beta (2019). Lithium-Ion Batteries. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1007/978-3-030-16800-1](https://doi.org/10.1007%2F978-3-030-16800-1). [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-3-030-16799-8](https://en.wikipedia.org/wiki/Special:BookSources/978-3-030-16799-8 "Special:BookSources/978-3-030-16799-8") . [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [155818532](https://api.semanticscholar.org/CorpusID:155818532). 43. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-43) ["Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube"](https://ghostarchive.org/varchive/youtube/20211030/7dtl650D0y0). www.youtube.com. 11 April 2019. Archived from [the original](https://www.youtube.com/watch?v=7dtl650D0y0) on 2021-10-30. Retrieved 2021-01-11. 44. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-44) Robertson, Adi (2022-04-06). ["OpenAI's DALL-E AI image generator can now edit pictures, too"](https://www.theverge.com/2022/4/6/23012123/openai-clip-dalle-2-ai-text-to-image-generator-testing). The Verge. Retrieved 2022-06-07. 45. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-45) ["The Stanford Natural Language Processing Group"](https://nlp.stanford.edu/projects/text2scene.shtml). nlp.stanford.edu. Retrieved 2022-06-07. 46. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-46) Coyne, Bob; Sproat, Richard (2001-08-01). ["WordsEye"](https://doi.org/10.1145/383259.383316). Proceedings of the 28th annual conference on Computer graphics and interactive techniques. SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. 487–496\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1145/383259.383316](https://doi.org/10.1145%2F383259.383316). [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1-58113-374-5](https://en.wikipedia.org/wiki/Special:BookSources/978-1-58113-374-5 "Special:BookSources/978-1-58113-374-5") . [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [3842372](https://api.semanticscholar.org/CorpusID:3842372). 47. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-47) ["Google announces AI advances in text-to-video, language translation, more"](https://venturebeat.com/ai/google-announces-ai-advances-in-text-to-video-language-translation-more/). VentureBeat. 2022-11-02. Retrieved 2022-11-09. 48. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-48) Vincent, James (2022-09-29). ["Meta's new text-to-video AI generator is like DALL-E for video"](https://www.theverge.com/2022/9/29/23378210/meta-text-to-video-ai-generation-make-a-video-model-dall-e). The Verge. Retrieved 2022-11-09. 49. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-49) ["Previous shared tasks \\| CoNLL"](https://www.conll.org/previous-tasks). www.conll.org. Retrieved 2021-01-11. 50. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-50) ["Cognition"](https://web.archive.org/web/20200715113427/https://www.lexico.com/definition/cognition). Lexico. [Oxford University Press](https://en.wikipedia.org/wiki/Oxford_University_Press "Oxford University Press") and [Dictionary.com](https://en.wikipedia.org/wiki/Dictionary.com "Dictionary.com"). Archived from [the original](https://www.lexico.com/definition/cognition) on July 15, 2020. Retrieved 6 May 2020. 51. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-51) ["Ask the Cognitive Scientist"](http://www.aft.org/newspubs/periodicals/ae/summer2002/willingham.cfm). American Federation of Teachers. 8 August 2014. "Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind." 52. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-52) Robinson, Peter (2008). Handbook of Cognitive Linguistics and Second Language Acquisition. Routledge. pp. 3–8\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-805-85352-0](https://en.wikipedia.org/wiki/Special:BookSources/978-0-805-85352-0 "Special:BookSources/978-0-805-85352-0") . 53. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-53) Lakoff, George (1999). Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm. New York Basic Books. pp. 569–583\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-465-05674-3](https://en.wikipedia.org/wiki/Special:BookSources/978-0-465-05674-3 "Special:BookSources/978-0-465-05674-3") . 54. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-54) Strauss, Claudia (1999). A Cognitive Theory of Cultural Meaning. Cambridge University Press. pp. 156–164\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-521-59541-4](https://en.wikipedia.org/wiki/Special:BookSources/978-0-521-59541-4 "Special:BookSources/978-0-521-59541-4") . 55. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-55) [US patent 9269353](https://worldwide.espacenet.com/textdoc?DB=EPODOC&IDX=US9269353) 56. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-56) ["Universal Conceptual Cognitive Annotation (UCCA)"](https://universalconceptualcognitiveannotation.github.io/). Universal Conceptual Cognitive Annotation (UCCA). Retrieved 2021-01-11. 57. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-57) Rodríguez, F. C., & Mairal-Usón, R. (2016). [Building an RRG computational grammar](https://www.redalyc.org/pdf/1345/134549291020.pdf). Onomazein, (34), 86–117. 58. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-58) ["Fluid Construction Grammar – A fully operational processing system for construction grammars"](https://www.fcg-net.org/). Retrieved 2021-01-11. 59. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-59) ["ACL Member Portal \\| The Association for Computational Linguistics Member Portal"](https://www.aclweb.org/portal/). www.aclweb.org. Retrieved 2021-01-11. 60. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-60) ["Chunks and Rules"](https://www.w3.org/Data/demos/chunks/chunks.html). W3C. Retrieved 2021-01-11. 61. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-61) Socher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014). ["Grounded Compositional Semantics for Finding and Describing Images with Sentences"](https://doi.org/10.1162%2Ftacl_a_00177). Transactions of the Association for Computational Linguistics. 2: 207–218\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1162/tacl\_a\_00177](https://doi.org/10.1162%2Ftacl_a_00177). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [2317858](https://api.semanticscholar.org/CorpusID:2317858). 62. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-62) Dasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). "Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al". [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[2207\.07051](https://arxiv.org/abs/2207.07051) \[[cs.CL](https://arxiv.org/archive/cs.CL)\]. 63. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-63) Friston, Karl J. (2022). Active Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference. The MIT Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-36997-8](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-36997-8 "Special:BookSources/978-0-262-36997-8") . ## Further reading \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=19 "Edit section: Further reading")\] - Bates, M (1995). ["Models of natural language understanding"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC40721). Proceedings of the National Academy of Sciences of the United States of America. 92 (22): 9977–9982\. [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[1995PNAS...92.9977B](https://ui.adsabs.harvard.edu/abs/1995PNAS...92.9977B). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1073/pnas.92.22.9977](https://doi.org/10.1073%2Fpnas.92.22.9977). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [40721](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC40721). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [7479812](https://pubmed.ncbi.nlm.nih.gov/7479812). - Steven Bird, Ewan Klein, and Edward Loper (2009). Natural Language Processing with Python. O'Reilly Media. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-596-51649-9](https://en.wikipedia.org/wiki/Special:BookSources/978-0-596-51649-9 "Special:BookSources/978-0-596-51649-9") . - [Kenna Hughes-Castleberry](https://en.wikipedia.org/w/index.php?title=Kenna_Hughes-Castleberry&action=edit&redlink=1 "Kenna Hughes-Castleberry (page does not exist)"), "A Murder Mystery Puzzle: The literary puzzle [Cain's Jawbone](https://en.wikipedia.org/wiki/Cain%27s_Jawbone "Cain's Jawbone"), which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms", [Scientific American](https://en.wikipedia.org/wiki/Scientific_American "Scientific American"), vol. 329, no. 4 (November 2023), pp. 81–82. "This murder mystery competition has revealed that although NLP ([natural-language processing](https://en.wikipedia.org/wiki/Natural-language_processing "Natural-language processing")) models are capable of incredible feats, their abilities are very much limited by the amount of [context](https://en.wikipedia.org/wiki/Context_\(linguistics\) "Context (linguistics)") they receive. This \[...\] could cause \[difficulties\] for researchers who hope to use them to do things such as analyze [ancient languages](https://en.wikipedia.org/wiki/Ancient_language "Ancient language"). In some cases, there are few historical records on long-gone [civilizations](https://en.wikipedia.org/wiki/Civilization "Civilization") to serve as [training data](https://en.wikipedia.org/wiki/Training_data "Training data") for such a purpose." (p. 82.) - Daniel Jurafsky and James H. Martin (2008). Speech and Language Processing, 2nd edition. Pearson Prentice Hall. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-13-187321-6](https://en.wikipedia.org/wiki/Special:BookSources/978-0-13-187321-6 "Special:BookSources/978-0-13-187321-6") . - Mohamed Zakaria Kurdi (2016). Natural Language Processing and Computational Linguistics: speech, morphology, and syntax, Volume 1. ISTE-Wiley. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1848218482](https://en.wikipedia.org/wiki/Special:BookSources/978-1848218482 "Special:BookSources/978-1848218482") . - Mohamed Zakaria Kurdi (2017). Natural Language Processing and Computational Linguistics: semantics, discourse, and applications, Volume 2. ISTE-Wiley. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1848219212](https://en.wikipedia.org/wiki/Special:BookSources/978-1848219212 "Special:BookSources/978-1848219212") . - Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008). Introduction to Information Retrieval. Cambridge University Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-521-86571-5](https://en.wikipedia.org/wiki/Special:BookSources/978-0-521-86571-5 "Special:BookSources/978-0-521-86571-5") . [Official html and pdf versions available without charge.](http://nlp.stanford.edu/IR-book/) - Christopher D. Manning and Hinrich Schütze (1999). Foundations of Statistical Natural Language Processing. The MIT Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-13360-9](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-13360-9 "Special:BookSources/978-0-262-13360-9") . - David M. W. Powers and Christopher C. R. Turk (1989). Machine Learning of Natural Language. Springer-Verlag. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-387-19557-5](https://en.wikipedia.org/wiki/Special:BookSources/978-0-387-19557-5 "Special:BookSources/978-0-387-19557-5") . ## External links \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=20 "Edit section: External links")\] - [![Wikimedia Commons logo](https://upload.wikimedia.org/wikipedia/en/thumb/4/4a/Commons-logo.svg/20px-Commons-logo.svg.png)](https://en.wikipedia.org/wiki/File:Commons-logo.svg) Media related to [Natural language processing](https://commons.wikimedia.org/wiki/Category:Natural_language_processing "commons:Category:Natural language processing") at Wikimedia Commons \| [v](https://en.wikipedia.org/wiki/Template:Natural_language_processing "Template:Natural language processing") [t](https://en.wikipedia.org/wiki/Template_talk:Natural_language_processing "Template talk:Natural language processing") [e](https://en.wikipedia.org/wiki/Special:EditPage/Template:Natural_language_processing "Special:EditPage/Template:Natural language processing")[Natural language processing]() \| \| \|---\|---\| \| General terms \| [AI-complete](https://en.wikipedia.org/wiki/AI-complete "AI-complete") [Bag-of-words](https://en.wikipedia.org/wiki/Bag-of-words_model "Bag-of-words model") [n\-gram](https://en.wikipedia.org/wiki/N-gram "N-gram") [Bigram](https://en.wikipedia.org/wiki/Bigram "Bigram") [Trigram](https://en.wikipedia.org/wiki/Trigram "Trigram") [Computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics") [Natural language understanding](https://en.wikipedia.org/wiki/Natural_language_understanding "Natural language understanding") [Stop words](https://en.wikipedia.org/wiki/Stop_word "Stop word") [Text processing](https://en.wikipedia.org/wiki/Text_processing "Text processing") \| \| [Text analysis](https://en.wikipedia.org/wiki/Text_mining "Text mining") \| \| \| \| \| \| [Text segmentation](https://en.wikipedia.org/wiki/Text_segmentation "Text segmentation") \| [Compound-term processing](https://en.wikipedia.org/wiki/Compound-term_processing "Compound-term processing") [Lemmatization](https://en.wikipedia.org/wiki/Lemmatization "Lemmatization") [Lexical analysis](https://en.wikipedia.org/wiki/Lexical_analysis "Lexical analysis") [Text chunking](https://en.wikipedia.org/wiki/Shallow_parsing "Shallow parsing") [Stemming](https://en.wikipedia.org/wiki/Stemming "Stemming") [Sentence segmentation](https://en.wikipedia.org/wiki/Sentence_boundary_disambiguation "Sentence boundary disambiguation") [Word segmentation](https://en.wikipedia.org/wiki/Word#Word_boundaries "Word") \| \| [Automatic summarization](https://en.wikipedia.org/wiki/Automatic_summarization "Automatic summarization") \| [Multi-document summarization](https://en.wikipedia.org/wiki/Multi-document_summarization "Multi-document summarization") [Sentence extraction](https://en.wikipedia.org/wiki/Sentence_extraction "Sentence extraction") [Text simplification](https://en.wikipedia.org/wiki/Text_simplification "Text simplification") \| \| [Machine translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation") \| [Computer-assisted](https://en.wikipedia.org/wiki/Computer-assisted_translation "Computer-assisted translation") [Example-based](https://en.wikipedia.org/wiki/Example-based_machine_translation "Example-based machine translation") [Rule-based](https://en.wikipedia.org/wiki/Rule-based_machine_translation "Rule-based machine translation") [Statistical](https://en.wikipedia.org/wiki/Statistical_machine_translation "Statistical machine translation") [Transfer-based](https://en.wikipedia.org/wiki/Transfer-based_machine_translation "Transfer-based machine translation") [Neural](https://en.wikipedia.org/wiki/Neural_machine_translation "Neural machine translation") \| \| [Distributional semantics](https://en.wikipedia.org/wiki/Distributional_semantics "Distributional semantics") models \| [BERT](https://en.wikipedia.org/wiki/BERT_\(language_model\) "BERT (language model)") [Document-term matrix](https://en.wikipedia.org/wiki/Document-term_matrix "Document-term matrix") [Explicit semantic analysis](https://en.wikipedia.org/wiki/Explicit_semantic_analysis "Explicit semantic analysis") [fastText](https://en.wikipedia.org/wiki/FastText "FastText") [GloVe](https://en.wikipedia.org/wiki/GloVe "GloVe") [Language model](https://en.wikipedia.org/wiki/Language_model "Language model") [large](https://en.wikipedia.org/wiki/Large_language_model "Large language model") [small](https://en.wikipedia.org/wiki/Small_language_model "Small language model") [Latent semantic analysis](https://en.wikipedia.org/wiki/Latent_semantic_analysis "Latent semantic analysis") [Long short-term memory](https://en.wikipedia.org/wiki/Long_short-term_memory "Long short-term memory") [Seq2seq](https://en.wikipedia.org/wiki/Seq2seq "Seq2seq") [Transformer](https://en.wikipedia.org/wiki/Transformer_\(deep_learning_architecture\) "Transformer (deep learning architecture)") [Word embedding](https://en.wikipedia.org/wiki/Word_embedding "Word embedding") [Word2vec](https://en.wikipedia.org/wiki/Word2vec "Word2vec") \| \| [Language resources](https://en.wikipedia.org/wiki/Language_resource "Language resource"), datasets and corpora \| \| \| \| \| \| Types and standards \| [Corpus linguistics](https://en.wikipedia.org/wiki/Corpus_linguistics "Corpus linguistics") [Lexical resource](https://en.wikipedia.org/wiki/Lexical_resource "Lexical resource") [Linguistic Linked Open Data](https://en.wikipedia.org/wiki/Linguistic_Linked_Open_Data "Linguistic Linked Open Data") [Machine-readable dictionary](https://en.wikipedia.org/wiki/Machine-readable_dictionary "Machine-readable dictionary") [Parallel text](https://en.wikipedia.org/wiki/Parallel_text "Parallel text") [PropBank](https://en.wikipedia.org/wiki/PropBank "PropBank") [Semantic network](https://en.wikipedia.org/wiki/Semantic_network "Semantic network") [Simple Knowledge Organization System](https://en.wikipedia.org/wiki/Simple_Knowledge_Organization_System "Simple Knowledge Organization System") [Speech corpus](https://en.wikipedia.org/wiki/Speech_corpus "Speech corpus") [Text corpus](https://en.wikipedia.org/wiki/Text_corpus "Text corpus") [Thesaurus (information retrieval)](https://en.wikipedia.org/wiki/Thesaurus_\(information_retrieval\) "Thesaurus (information retrieval)") [Treebank](https://en.wikipedia.org/wiki/Treebank "Treebank") [Universal Dependencies](https://en.wikipedia.org/wiki/Universal_Dependencies "Universal Dependencies") \| \| Data \| [BabelNet](https://en.wikipedia.org/wiki/BabelNet "BabelNet") [Bank of English](https://en.wikipedia.org/wiki/Bank_of_English "Bank of English") [DBpedia](https://en.wikipedia.org/wiki/DBpedia "DBpedia") [FrameNet](https://en.wikipedia.org/wiki/FrameNet "FrameNet") [Google Ngram Viewer](https://en.wikipedia.org/wiki/Google_Ngram_Viewer "Google Ngram Viewer") [UBY](https://en.wikipedia.org/wiki/UBY "UBY") [WordNet](https://en.wikipedia.org/wiki/WordNet "WordNet") [Wikidata](https://en.wikipedia.org/wiki/Wikidata "Wikidata") \| \| [Automatic identification and data capture](https://en.wikipedia.org/wiki/Automatic_identification_and_data_capture "Automatic identification and data capture") \| [Speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition") [Speech segmentation](https://en.wikipedia.org/wiki/Speech_segmentation "Speech segmentation") [Speech synthesis](https://en.wikipedia.org/wiki/Speech_synthesis "Speech synthesis") [Natural language generation](https://en.wikipedia.org/wiki/Natural_language_generation "Natural language generation") [Optical character recognition](https://en.wikipedia.org/wiki/Optical_character_recognition "Optical character recognition") \| \| [Topic model](https://en.wikipedia.org/wiki/Topic_model "Topic model") \| [Document classification](https://en.wikipedia.org/wiki/Document_classification "Document classification") [Latent Dirichlet allocation](https://en.wikipedia.org/wiki/Latent_Dirichlet_allocation "Latent Dirichlet allocation") [Pachinko allocation](https://en.wikipedia.org/wiki/Pachinko_allocation "Pachinko allocation") \| \| [Computer-assisted reviewing](https://en.wikipedia.org/wiki/Computer-assisted_reviewing "Computer-assisted reviewing") \| [Automated essay scoring](https://en.wikipedia.org/wiki/Automated_essay_scoring "Automated essay scoring") [Concordancer](https://en.wikipedia.org/wiki/Concordancer "Concordancer") [Grammar checker](https://en.wikipedia.org/wiki/Grammar_checker "Grammar checker") [Predictive text](https://en.wikipedia.org/wiki/Predictive_text "Predictive text") [Pronunciation assessment](https://en.wikipedia.org/wiki/Pronunciation_assessment "Pronunciation assessment") [Spell checker](https://en.wikipedia.org/wiki/Spell_checker "Spell checker") \| \| [Natural language user interface](https://en.wikipedia.org/wiki/Natural-language_user_interface "Natural-language user interface") \| [Chatbot](https://en.wikipedia.org/wiki/Chatbot "Chatbot") [Interactive fiction](https://en.wikipedia.org/wiki/Interactive_fiction "Interactive fiction") [Question answering](https://en.wikipedia.org/wiki/Question_answering "Question answering") [Virtual assistant](https://en.wikipedia.org/wiki/Virtual_assistant "Virtual assistant") [Voice user interface](https://en.wikipedia.org/wiki/Voice_user_interface "Voice user interface") \| \| Related \| [Formal semantics](https://en.wikipedia.org/wiki/Formal_semantics_\(natural_language\) "Formal semantics (natural language)") [Gensim](https://en.wikipedia.org/wiki/Gensim "Gensim") [Hallucination](https://en.wikipedia.org/wiki/Hallucination_\(artificial_intelligence\) "Hallucination (artificial intelligence)") [Natural Language Toolkit](https://en.wikipedia.org/wiki/Natural_Language_Toolkit "Natural Language Toolkit") [spaCy](https://en.wikipedia.org/wiki/SpaCy "SpaCy") \| [Portal](https://en.wikipedia.org/wiki/Wikipedia:Contents/Portals "Wikipedia:Contents/Portals"): - [![icon](https://upload.wikimedia.org/wikipedia/commons/thumb/d/de/Globe_of_letters.svg/20px-Globe_of_letters.svg.png)](https://en.wikipedia.org/wiki/File:Globe_of_letters.svg) [Language](https://en.wikipedia.org/wiki/Portal:Language "Portal:Language") \| [Authority control databases](https://en.wikipedia.org/wiki/Help:Authority_control "Help:Authority control") [![Edit this at Wikidata](https://upload.wikimedia.org/wikipedia/en/thumb/8/8a/OOjs_UI_icon_edit-ltr-progressive.svg/20px-OOjs_UI_icon_edit-ltr-progressive.svg.png)](https://www.wikidata.org/wiki/Q30642#identifiers "Edit this at Wikidata") \| \| \|---\|---\| \| National \| [United States](https://id.loc.gov/authorities/sh88002425) [Japan](https://id.ndl.go.jp/auth/ndlna/00562347) [Czech Republic](https://aleph.nkp.cz/F/?func=find-c&local_base=aut&ccl_term=ica=ph427562&CON_LNG=ENG) [Israel](https://www.nli.org.il/en/authorities/987007536703305171) \| \| Other \| [Yale LUX](https://lux.collections.yale.edu/view/concept/e0fc4a1c-7602-43f9-9cea-0d8bb49039b8) \| ![](https://en.wikipedia.org/wiki/Special:CentralAutoLogin/start?useformat=desktop&type=1x1&usesul3=1) Retrieved from "<https://en.wikipedia.org/w/index.php?title=Natural_language_processing&oldid=1343173705>" [Categories](https://en.wikipedia.org/wiki/Help:Category "Help:Category"): - [Natural language processing](https://en.wikipedia.org/wiki/Category:Natural_language_processing "Category:Natural language processing") - [Computational fields of study](https://en.wikipedia.org/wiki/Category:Computational_fields_of_study "Category:Computational fields of study") - [Computational linguistics](https://en.wikipedia.org/wiki/Category:Computational_linguistics "Category:Computational linguistics") - [Speech recognition](https://en.wikipedia.org/wiki/Category:Speech_recognition "Category:Speech recognition") Hidden categories: - [All accuracy disputes](https://en.wikipedia.org/wiki/Category:All_accuracy_disputes "Category:All accuracy disputes") - [Accuracy disputes from December 2013](https://en.wikipedia.org/wiki/Category:Accuracy_disputes_from_December_2013 "Category:Accuracy disputes from December 2013") - [Harv and Sfn no-target errors](https://en.wikipedia.org/wiki/Category:Harv_and_Sfn_no-target_errors "Category:Harv and Sfn no-target errors") - [CS1 errors: periodical ignored](https://en.wikipedia.org/wiki/Category:CS1_errors:_periodical_ignored "Category:CS1 errors: periodical ignored") - [CS1 maint: work parameter with ISBN](https://en.wikipedia.org/wiki/Category:CS1_maint:_work_parameter_with_ISBN "Category:CS1 maint: work parameter with ISBN") - [CS1 maint: location](https://en.wikipedia.org/wiki/Category:CS1_maint:_location "Category:CS1 maint: location") - [Articles with short description](https://en.wikipedia.org/wiki/Category:Articles_with_short_description "Category:Articles with short description") - [Short description is different from Wikidata](https://en.wikipedia.org/wiki/Category:Short_description_is_different_from_Wikidata "Category:Short description is different from Wikidata") - [Articles needing additional references from May 2024](https://en.wikipedia.org/wiki/Category:Articles_needing_additional_references_from_May_2024 "Category:Articles needing additional references from May 2024") - [All articles needing additional references](https://en.wikipedia.org/wiki/Category:All_articles_needing_additional_references "Category:All articles needing additional references") - [Wikipedia articles needing rewrite from July 2025](https://en.wikipedia.org/wiki/Category:Wikipedia_articles_needing_rewrite_from_July_2025 "Category:Wikipedia articles needing rewrite from July 2025") - [All articles needing rewrite](https://en.wikipedia.org/wiki/Category:All_articles_needing_rewrite "Category:All articles needing rewrite") - [Wikipedia articles needing reorganization from July 2025](https://en.wikipedia.org/wiki/Category:Wikipedia_articles_needing_reorganization_from_July_2025 "Category:Wikipedia articles needing reorganization from July 2025") - [Articles with multiple maintenance issues](https://en.wikipedia.org/wiki/Category:Articles_with_multiple_maintenance_issues "Category:Articles with multiple maintenance issues") - [Commons category link from Wikidata](https://en.wikipedia.org/wiki/Category:Commons_category_link_from_Wikidata "Category:Commons category link from Wikidata") - This page was last edited on 12 March 2026, at 18:37 (UTC). - Text is available under the [Creative Commons Attribution-ShareAlike 4.0 License](https://en.wikipedia.org/wiki/Wikipedia:Text_of_the_Creative_Commons_Attribution-ShareAlike_4.0_International_License "Wikipedia:Text of the Creative Commons Attribution-ShareAlike 4.0 International License"); additional terms may apply. By using this site, you agree to the [Terms of Use](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Terms_of_Use "foundation:Special:MyLanguage/Policy:Terms of Use") and [Privacy Policy](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Privacy_policy "foundation:Special:MyLanguage/Policy:Privacy policy"). Wikipedia® is a registered trademark of the [Wikimedia Foundation, Inc.](https://wikimediafoundation.org/), a non-profit organization. - [Privacy policy](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Privacy_policy) - [About Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:About) - [Disclaimers](https://en.wikipedia.org/wiki/Wikipedia:General_disclaimer) - [Contact Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:Contact_us) - [Legal & safety contacts](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Legal:Wikimedia_Foundation_Legal_and_Safety_Contact_Information) - [Code of Conduct](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Universal_Code_of_Conduct) - [Developers](https://developer.wikimedia.org/) - [Statistics](https://stats.wikimedia.org/#/en.wikipedia.org) - [Cookie statement](https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Cookie_statement) - [Mobile view](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&mobileaction=toggle_view_mobile) - [![Wikimedia Foundation](https://en.wikipedia.org/static/images/footer/wikimedia.svg)](https://www.wikimedia.org/) - [![Powered by MediaWiki](https://en.wikipedia.org/w/resources/assets/mediawiki_compact.svg)](https://www.mediawiki.org/) Search Toggle the table of contents Natural language processing 71 languages [Add topic](https://en.wikipedia.org/wiki/Natural_language_processing)
Readable Markdown	From Wikipedia, the free encyclopedia Natural language processing (NLP) is the processing of [natural language](https://en.wikipedia.org/wiki/Natural_language "Natural language") information by a [computer](https://en.wikipedia.org/wiki/Computer "Computer"). NLP is a subfield of [computer science](https://en.wikipedia.org/wiki/Computer_science "Computer science") and is closely associated with [artificial intelligence](https://en.wikipedia.org/wiki/Artificial_intelligence "Artificial intelligence"). NLP is also related to [information retrieval](https://en.wikipedia.org/wiki/Information_retrieval "Information retrieval"), [knowledge representation](https://en.wikipedia.org/wiki/Knowledge_representation_and_reasoning "Knowledge representation and reasoning"), [computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics"), and [linguistics](https://en.wikipedia.org/wiki/Linguistics "Linguistics") more broadly.[\[1\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-nlpintro-1) Major processing tasks in an NLP system include: [speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition"), [text classification](https://en.wikipedia.org/wiki/Text_classification "Text classification"), [natural language understanding](https://en.wikipedia.org/wiki/Natural_language_understanding "Natural language understanding"), and [natural language generation](https://en.wikipedia.org/wiki/Natural_language_generation "Natural language generation"). Natural language processing has its roots in the 1950s.[\[2\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-2) Already in 1950, [Alan Turing](https://en.wikipedia.org/wiki/Alan_Turing "Alan Turing") published an article titled "[Computing Machinery and Intelligence](https://en.wikipedia.org/wiki/Computing_Machinery_and_Intelligence "Computing Machinery and Intelligence")" which proposed what is now called the [Turing test](https://en.wikipedia.org/wiki/Turing_test "Turing test") as a criterion of intelligence, though at the time that was not articulated as a problem separate from artificial intelligence. The proposed test includes a task that involves the automated interpretation and generation of natural language. ### Symbolic NLP (1950s – early 1990s) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=2 "Edit section: Symbolic NLP (1950s – early 1990s)")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/0/09/AST_Document.svg/250px-AST_Document.svg.png)](https://en.wikipedia.org/wiki/File:AST_Document.svg) A document parsed into an abstract syntax tree The premise of symbolic NLP is often illustrated using [John Searle's Chinese room](https://en.wikipedia.org/wiki/John_Searle_\(American_philosopher\) "John Searle (American philosopher)") thought experiment: Given a collection of rules (e.g., a Chinese phrasebook, with questions and matching answers), the computer emulates natural language understanding (or other NLP tasks) by applying those rules to the data it confronts. - 1950s: The [Georgetown experiment](https://en.wikipedia.org/wiki/Georgetown-IBM_experiment "Georgetown-IBM experiment") in 1954 involved fully [automatic translation](https://en.wikipedia.org/wiki/Automatic_translation "Automatic translation") of more than sixty Russian sentences into English. The authors claimed that within three or five years, machine translation would be a solved problem.[\[3\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-3) However, real progress was much slower, and after the [ALPAC report](https://en.wikipedia.org/wiki/ALPAC "ALPAC") in 1966, which found that ten years of research had failed to fulfill the expectations, funding for machine translation was dramatically reduced. Little further research in machine translation was conducted in America (though some research continued elsewhere, such as Japan and Europe[\[4\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-4)) until the late 1980s when the first [statistical machine translation](https://en.wikipedia.org/wiki/Statistical_machine_translation "Statistical machine translation") systems were developed. - 1960s: Some notably successful natural language processing systems developed in the 1960s were [SHRDLU](https://en.wikipedia.org/wiki/SHRDLU "SHRDLU"), a natural language system working in restricted "[blocks worlds](https://en.wikipedia.org/wiki/Blocks_world "Blocks world")" with restricted vocabularies, and [ELIZA](https://en.wikipedia.org/wiki/ELIZA "ELIZA"), a simulation of a [Rogerian psychotherapy](https://en.wikipedia.org/wiki/Rogerian_psychotherapy "Rogerian psychotherapy"), written by [Joseph Weizenbaum](https://en.wikipedia.org/wiki/Joseph_Weizenbaum "Joseph Weizenbaum") between 1964 and 1966. Despite using minimal information about human thought or emotion, ELIZA was able to produce interactions that appeared human-like. When the "patient" exceeded the very small knowledge base, ELIZA might provide a generic response, for example, responding to "My head hurts" with "Why do you say your head hurts?". Ross Quillian's successful work on natural language was demonstrated with a vocabulary of only twenty words, because that was all that would fit in a computer memory at the time.[\[5\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-5) - 1970s: During the 1970s, many programmers began to write "conceptual [ontologies](https://en.wikipedia.org/wiki/Ontology_\(information_science\) "Ontology (information science)")", which structured real-world information into computer-understandable data. Examples are MARGIE (Schank, 1975), SAM (Cullingford, 1978), PAM (Wilensky, 1978), TaleSpin (Meehan, 1976), QUALM (Lehnert, 1977), Politics (Carbonell, 1979), and Plot Units (Lehnert 1981). During this time, the first [chatterbots](https://en.wikipedia.org/wiki/Chatterbots "Chatterbots") were written (e.g., [PARRY](https://en.wikipedia.org/wiki/PARRY "PARRY")). - 1980s: The 1980s and early 1990s mark the heyday of symbolic methods in NLP. Focus areas of the time included research on rule-based parsing (e.g., the development of [HPSG](https://en.wikipedia.org/wiki/Head-driven_phrase_structure_grammar "Head-driven phrase structure grammar") as a computational operationalization of [generative grammar](https://en.wikipedia.org/wiki/Generative_grammar "Generative grammar")), morphology (e.g., two-level morphology[\[6\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-6)), semantics (e.g., [Lesk algorithm](https://en.wikipedia.org/wiki/Lesk_algorithm "Lesk algorithm")), reference (e.g., within Centering Theory[\[7\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-7)) and other areas of natural language understanding (e.g., in the [Rhetorical Structure Theory](https://en.wikipedia.org/wiki/Rhetorical_structure_theory "Rhetorical structure theory")). Other lines of research were continued, e.g., the development of chatterbots with [Racter](https://en.wikipedia.org/wiki/Racter "Racter") and [Jabberwacky](https://en.wikipedia.org/wiki/Jabberwacky "Jabberwacky"). An important development (that eventually led to the statistical turn in the 1990s) was the rising importance of quantitative evaluation in this period.[\[8\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-8) ### Statistical NLP (1990s–present) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=3 "Edit section: Statistical NLP (1990s–present)")\] Up until the 1980s, most natural language processing systems were based on complex sets of hand-written rules. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of [machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning") algorithms for language processing. This shift was influenced by increasing computational power (see [Moore's law](https://en.wikipedia.org/wiki/Moore%27s_law "Moore's law")) and a decline in the dominance of [Chomskyan](https://en.wikipedia.org/wiki/Chomskyan "Chomskyan") linguistic theories... (e.g. [transformational grammar](https://en.wikipedia.org/wiki/Transformational_grammar "Transformational grammar")), whose theoretical underpinnings discouraged the sort of [corpus linguistics](https://en.wikipedia.org/wiki/Corpus_linguistics "Corpus linguistics") that underlies the machine-learning approach to language processing.[\[9\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-9) - 1990s: Many of the notable early successes in statistical methods in NLP occurred in the field of [machine translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation"), due especially to work at IBM Research, such as [IBM alignment models](https://en.wikipedia.org/wiki/IBM_alignment_models "IBM alignment models"). These systems were able to take advantage of existing multilingual [textual corpora](https://en.wikipedia.org/wiki/Text_corpus "Text corpus") that had been produced by the [Parliament of Canada](https://en.wikipedia.org/wiki/Parliament_of_Canada "Parliament of Canada") and the [European Union](https://en.wikipedia.org/wiki/European_Union "European Union") as a result of laws calling for the translation of all governmental proceedings into all official languages of the corresponding systems of government. However, many systems relied on corpora that were specifically developed for the tasks they were designed to perform. This reliance has been a major limitation to their broader effectiveness and continues to affect similar systems. Consequently, significant research has focused on methods for learning effectively from limited amounts of data. - 2000s: With the growth of the web, increasing amounts of raw (unannotated) language data have become available since the mid-1990s. Research has thus increasingly focused on [unsupervised](https://en.wikipedia.org/wiki/Unsupervised_learning "Unsupervised learning") and [semi-supervised learning](https://en.wikipedia.org/wiki/Semi-supervised_learning "Semi-supervised learning") algorithms. Such algorithms can learn from data that has not been hand-annotated with the desired answers or using a combination of annotated and non-annotated data. Generally, this task is much more difficult than [supervised learning](https://en.wikipedia.org/wiki/Supervised_learning "Supervised learning"), and typically produces less accurate results for a given amount of input data. However, large quantities of non-annotated data are available (including, among other things, the entire content of the [World Wide Web](https://en.wikipedia.org/wiki/World_Wide_Web "World Wide Web")), which can often make up for the worse efficiency if the algorithm used has a low enough [time complexity](https://en.wikipedia.org/wiki/Time_complexity "Time complexity") to be practical. - 2003: [word n-gram model](https://en.wikipedia.org/wiki/Word_n-gram_language_model "Word n-gram language model"), at the time the best statistical algorithm, is outperformed by a [multi-layer perceptron](https://en.wikipedia.org/wiki/Multi-layer_perceptron "Multi-layer perceptron") (with a single hidden layer and context length of several words, trained on up to 14 million words, by [Bengio](https://en.wikipedia.org/wiki/Yoshua_Bengio "Yoshua Bengio") et al.)[\[10\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-10) - 2010: [Tomáš Mikolov](https://en.wikipedia.org/wiki/Tom%C3%A1%C5%A1_Mikolov "Tomáš Mikolov") (then a PhD student at [Brno University of Technology](https://en.wikipedia.org/wiki/Brno_University_of_Technology "Brno University of Technology")) with co-authors applied a simple [recurrent neural network](https://en.wikipedia.org/wiki/Recurrent_neural_network "Recurrent neural network") with a single hidden layer to language modeling,[\[11\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-11) and in the following years he went on to develop [Word2vec](https://en.wikipedia.org/wiki/Word2vec "Word2vec"). In the 2010s, [representation learning](https://en.wikipedia.org/wiki/Representation_learning "Representation learning") and [deep neural network](https://en.wikipedia.org/wiki/Deep_learning "Deep learning")\-style (featuring many hidden layers) machine learning methods became widespread in natural language processing. This shift gained momentum due to results showing that such techniques[\[12\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-goldberg:nnlp17-12)[\[13\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-goodfellow:book16-13) can achieve state-of-the-art results in many natural language tasks, e.g., in [language modeling](https://en.wikipedia.org/wiki/Language_modeling "Language modeling")[\[14\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-jozefowicz:lm16-14) and parsing.[\[15\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-choe:emnlp16-15)[\[16\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-vinyals:nips15-16) This is increasingly important [in medicine and healthcare](https://en.wikipedia.org/wiki/Artificial_intelligence_in_healthcare "Artificial intelligence in healthcare"), where NLP helps analyze notes and text in [electronic health records](https://en.wikipedia.org/wiki/Electronic_health_record "Electronic health record") that would otherwise be inaccessible for study when seeking to improve care[\[17\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-17) or protect patient privacy.[\[18\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-18) ## Approaches: Symbolic, statistical, neural networks \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=4 "Edit section: Approaches: Symbolic, statistical, neural networks")\] Symbolic approach, i.e., the hand-coding of a set of rules for manipulating symbols, coupled with a dictionary lookup, was historically the first approach used both by AI in general and by NLP in particular:[\[19\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-winograd:shrdlu71-19)[\[20\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-schank77-20) such as by writing grammars or devising heuristic rules for [stemming](https://en.wikipedia.org/wiki/Stemming "Stemming"). [Machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning") approaches, which include both statistical and neural networks, on the other hand, have many advantages over the symbolic approach: - both statistical and neural networks methods can focus more on the most common cases extracted from a corpus of texts, whereas the rule-based approach needs to provide rules for both rare cases and common ones equally. - [language models](https://en.wikipedia.org/wiki/Language_model "Language model"), produced by either statistical or neural networks methods, are more robust to both unfamiliar (e.g. containing words or structures that have not been seen before) and erroneous input (e.g. with misspelled words or words accidentally omitted) in comparison to the rule-based systems, which are also more costly to produce. - the larger such a (probabilistic) language model is, the more accurate it becomes, in contrast to rule-based systems that can gain accuracy only by increasing the amount and complexity of the rules leading to [intractability](https://en.wikipedia.org/wiki/Intractable_problem "Intractable problem") problems. Rule-based systems are commonly used: - when the amount of training data is insufficient to successfully apply machine learning methods, e.g., for the machine translation of low-resource languages such as provided by the [Apertium](https://en.wikipedia.org/wiki/Apertium "Apertium") system, - for preprocessing in NLP pipelines, e.g., [tokenization](https://en.wikipedia.org/wiki/Tokenization_\(lexical_analysis\) "Tokenization (lexical analysis)"), or - for post-processing and transforming the output of NLP pipelines, e.g., for [knowledge extraction](https://en.wikipedia.org/wiki/Knowledge_extraction "Knowledge extraction") from syntactic parses. ### Statistical approach \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=5 "Edit section: Statistical approach")\] In the late 1980s and mid-1990s, the statistical approach ended a period of [AI winter](https://en.wikipedia.org/wiki/AI_winter "AI winter"), which was caused by the inefficiencies of the rule-based approaches.[\[21\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-johnson:eacl:ilcl09-21)[\[22\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-resnik:langlog11-22) The earliest [decision trees](https://en.wikipedia.org/wiki/Decision_tree "Decision tree"), producing systems of hard [if–then rules](https://en.wikipedia.org/wiki/Conditional_\(computer_programming\)#If%E2%80%93then\(%E2%80%93else\) "Conditional (computer programming)"), were still very similar to the old rule-based approaches. Only the introduction of hidden [Markov models](https://en.wikipedia.org/wiki/Markov_model "Markov model"), applied to part-of-speech tagging, announced the end of the old rule-based approach. A major drawback of statistical methods is that they require elaborate [feature engineering](https://en.wikipedia.org/wiki/Feature_engineering "Feature engineering"). Since 2015,[\[23\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-23) [neural network](https://en.wikipedia.org/wiki/Neural_network_\(machine_learning\) "Neural network (machine learning)")–based methods have increasingly replaced traditional statistical approaches, using [semantic networks](https://en.wikipedia.org/wiki/Semantic_networks "Semantic networks")[\[24\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-24) and [word embeddings](https://en.wikipedia.org/wiki/Word_embedding "Word embedding") to capture semantic properties of words. Intermediate tasks (e.g., part-of-speech tagging and dependency parsing) are not needed anymore. [Neural machine translation](https://en.wikipedia.org/wiki/Neural_machine_translation "Neural machine translation"), based on then-newly invented [sequence-to-sequence](https://en.wikipedia.org/wiki/Seq2seq "Seq2seq") transformations, made obsolete the intermediate steps, such as word alignment, previously necessary for [statistical machine translation](https://en.wikipedia.org/wiki/Statistical_machine_translation "Statistical machine translation"). The following is a list of some of the most commonly researched tasks in natural language processing. Some of these tasks have direct real-world applications, while others more commonly serve as subtasks that are used to aid in solving larger tasks. Though natural language processing tasks are closely intertwined, they can be subdivided into categories for convenience. A coarse division is given below. ### Text and speech processing \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=8 "Edit section: Text and speech processing")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/8/8c/Hebrew_stop_words_word_cloud.png/250px-Hebrew_stop_words_word_cloud.png)](https://en.wikipedia.org/wiki/File:Hebrew_stop_words_word_cloud.png) Word cloud of stop words in Hebrew [Optical character recognition](https://en.wikipedia.org/wiki/Optical_character_recognition "Optical character recognition") (OCR) Given an image representing printed text, determine the corresponding text. [Speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition") Given a sound clip of a person or people speaking, determine the textual representation of the speech. This is the opposite of [text to speech](https://en.wikipedia.org/wiki/Text_to_speech "Text to speech") and is one of the extremely difficult problems colloquially termed "[AI-complete](https://en.wikipedia.org/wiki/AI-complete "AI-complete")" (see above). In [natural speech](https://en.wikipedia.org/wiki/Natural_speech "Natural speech") there are hardly any pauses between successive words, and thus [speech segmentation](https://en.wikipedia.org/wiki/Speech_segmentation "Speech segmentation") is a necessary subtask of speech recognition (see below). In most spoken languages, the sounds representing successive letters blend into each other in a process termed [coarticulation](https://en.wikipedia.org/wiki/Coarticulation "Coarticulation"), so the conversion of the [analog signal](https://en.wikipedia.org/wiki/Analog_signal "Analog signal") to discrete characters can be a very difficult process. Also, given that words in the same language are spoken by people with different accents, the speech recognition software must be able to recognize the wide variety of input as being identical to each other in terms of its textual equivalent. [Speech segmentation](https://en.wikipedia.org/wiki/Speech_segmentation "Speech segmentation") Given a sound clip of a person or people speaking, separate it into words. A subtask of [speech recognition](https://en.wikipedia.org/wiki/Speech_recognition "Speech recognition") and typically grouped with it. [Text-to-speech](https://en.wikipedia.org/wiki/Text-to-speech "Text-to-speech") Given a text, transform those units and produce a spoken representation. Text-to-speech can be used to aid the visually impaired.[\[25\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-25) [Word segmentation](https://en.wikipedia.org/wiki/Word_segmentation "Word segmentation") ([Tokenization](https://en.wikipedia.org/wiki/Tokenization_\(lexical_analysis\) "Tokenization (lexical analysis)")) [Tokenization](https://en.wikipedia.org/wiki/Tokenization_\(data_security\) "Tokenization (data security)") is a text-processing technique that divides text into individual words or word fragments. This technique results in two key components: a word index and tokenized text. The word index is a list that maps unique words to specific numerical identifiers, and the tokenized text replaces each word with its corresponding numerical token. These numerical tokens are then used in various deep learning methods.[\[26\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-:0-26) For a language like [English](https://en.wikipedia.org/wiki/English_language "English language"), this is fairly trivial, since words are usually separated by spaces. However, some written languages like [Chinese](https://en.wikipedia.org/wiki/Chinese_language "Chinese language"), [Japanese](https://en.wikipedia.org/wiki/Japanese_language "Japanese language") and [Thai](https://en.wikipedia.org/wiki/Thai_language "Thai language") do not mark word boundaries in such a fashion, and in those languages text segmentation is a significant task requiring knowledge of the [vocabulary](https://en.wikipedia.org/wiki/Vocabulary "Vocabulary") and [morphology](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") of words in the language. Sometimes this process is also used in cases like [bag of words](https://en.wikipedia.org/wiki/Bag_of_words "Bag of words") (BOW) creation in data mining.[\[27\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-27)[\[28\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-28) ### Morphological analysis \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=9 "Edit section: Morphological analysis")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/Eustagger_euskal_lematizatzailearen_adibidea.png/250px-Eustagger_euskal_lematizatzailearen_adibidea.png)](https://en.wikipedia.org/wiki/File:Eustagger_euskal_lematizatzailearen_adibidea.png) Lemmatization of Basque words [Lemmatization](https://en.wikipedia.org/wiki/Lemmatisation "Lemmatisation") The task of removing inflectional endings only and to return the base dictionary form of a word which is also known as a lemma. Lemmatization is another technique for reducing words to their normalized form. But in this case, the transformation actually uses a dictionary to map words to their actual form.[\[29\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-29) [Morphological segmentation](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") Separate words into individual [morphemes](https://en.wikipedia.org/wiki/Morpheme "Morpheme") and identify the class of the morphemes. The difficulty of this task depends greatly on the complexity of the [morphology](https://en.wikipedia.org/wiki/Morphology_\(linguistics\) "Morphology (linguistics)") (i.e., the structure of words) of the language being considered. [English](https://en.wikipedia.org/wiki/English_language "English language") has fairly simple morphology, especially [inflectional morphology](https://en.wikipedia.org/wiki/Inflectional_morphology "Inflectional morphology"), and thus it is often possible to ignore this task entirely and simply model all possible forms of a word (e.g., "open, opens, opened, opening") as separate words. In languages such as [Turkish](https://en.wikipedia.org/wiki/Turkish_language "Turkish language") or [Meitei](https://en.wikipedia.org/wiki/Meitei_language "Meitei language"), a highly [agglutinated](https://en.wikipedia.org/wiki/Agglutination "Agglutination") Indian language, however, such an approach is not possible, as each dictionary entry has thousands of possible word forms.[\[30\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-30) [Part-of-speech tagging](https://en.wikipedia.org/wiki/Part-of-speech_tagging "Part-of-speech tagging") Given a sentence, determine the [part of speech](https://en.wikipedia.org/wiki/Part_of_speech "Part of speech") (POS) for each word. Many words, especially common ones, can serve as multiple parts of speech. For example, "book" can be a [noun](https://en.wikipedia.org/wiki/Noun "Noun") ("the book on the table") or [verb](https://en.wikipedia.org/wiki/Verb "Verb") ("to book a flight"); "set" can be a noun, verb or [adjective](https://en.wikipedia.org/wiki/Adjective "Adjective"); and "out" can be any of at least five different parts of speech. [Stemming](https://en.wikipedia.org/wiki/Stemming "Stemming") The process of reducing inflected (or sometimes derived) words to a base form (e.g., "close" will be the root for "closed", "closing", "close", "closer" etc.). Stemming yields similar results as lemmatization, but does so on grounds of rules, not a dictionary. [Grammar induction](https://en.wikipedia.org/wiki/Grammar_induction "Grammar induction")[\[31\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-31) Generate a [formal grammar](https://en.wikipedia.org/wiki/Formal_grammar "Formal grammar") that describes a language's syntax. [Sentence breaking](https://en.wikipedia.org/wiki/Sentence_breaking "Sentence breaking") (also known as "[sentence boundary disambiguation](https://en.wikipedia.org/wiki/Sentence_boundary_disambiguation "Sentence boundary disambiguation")") Given a chunk of text, find the sentence boundaries. Sentence boundaries are often marked by [periods](https://en.wikipedia.org/wiki/Full_stop "Full stop") or other [punctuation marks](https://en.wikipedia.org/wiki/Punctuation_mark "Punctuation mark"), but these same characters can serve other purposes (e.g., marking [abbreviations](https://en.wikipedia.org/wiki/Abbreviation "Abbreviation")). [Parsing](https://en.wikipedia.org/wiki/Parsing "Parsing") Determine the [parse tree](https://en.wikipedia.org/wiki/Parse_tree "Parse tree") (grammatical analysis) of a given sentence. The [grammar](https://en.wikipedia.org/wiki/Grammar "Grammar") for [natural languages](https://en.wikipedia.org/wiki/Natural_language "Natural language") is [ambiguous](https://en.wikipedia.org/wiki/Ambiguous "Ambiguous") and typical sentences have multiple possible analyses: perhaps surprisingly, for a typical sentence there may be thousands of potential parses (most of which will seem completely nonsensical to a human). There are two primary types of parsing: dependency parsing and constituency parsing. Dependency parsing focuses on the relationships between words in a sentence (marking things like primary objects and predicates), whereas constituency parsing focuses on building out the parse tree using a [probabilistic context-free grammar](https://en.wikipedia.org/wiki/Probabilistic_context-free_grammar "Probabilistic context-free grammar") (PCFG) (see also [stochastic grammar](https://en.wikipedia.org/wiki/Stochastic_grammar "Stochastic grammar")). ### Lexical semantics (of individual words in context) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=11 "Edit section: Lexical semantics (of individual words in context)")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/3/34/Entity_Linking_-_Example_of_pipeline.png/250px-Entity_Linking_-_Example_of_pipeline.png)](https://en.wikipedia.org/wiki/File:Entity_Linking_-_Example_of_pipeline.png) An entity linking pipeline [Lexical semantics](https://en.wikipedia.org/wiki/Lexical_semantics "Lexical semantics") What is the computational meaning of individual words in context? [Distributional semantics](https://en.wikipedia.org/wiki/Distributional_semantics "Distributional semantics") How can we learn semantic representations from data? [Named entity recognition](https://en.wikipedia.org/wiki/Named_entity_recognition "Named entity recognition") (NER) Given a stream of text, determine which items in the text map to proper names, such as people or places, and what the type of each such name is (e.g. person, location, organization). Although [capitalization](https://en.wikipedia.org/wiki/Capitalization "Capitalization") can aid in recognizing named entities in languages such as English, this information cannot aid in determining the type of [named entity](https://en.wikipedia.org/wiki/Named_entity "Named entity"), and in any case, is often inaccurate or insufficient. For example, the first letter of a sentence is also capitalized, and named entities often span several words, only some of which are capitalized. Furthermore, many other languages in non-Western scripts (e.g. [Chinese](https://en.wikipedia.org/wiki/Chinese_language "Chinese language") or [Arabic](https://en.wikipedia.org/wiki/Arabic_language "Arabic language")) do not have any capitalization at all, and even languages with capitalization may not consistently use it to distinguish names. For example, [German](https://en.wikipedia.org/wiki/German_language "German language") capitalizes all [nouns](https://en.wikipedia.org/wiki/Noun "Noun"), regardless of whether they are names, and [French](https://en.wikipedia.org/wiki/French_language "French language") and [Spanish](https://en.wikipedia.org/wiki/Spanish_language "Spanish language") do not capitalize names that serve as [adjectives](https://en.wikipedia.org/wiki/Adjective "Adjective"). This task is also referred to as token classification.[\[32\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-32) [Sentiment analysis](https://en.wikipedia.org/wiki/Sentiment_analysis "Sentiment analysis") (see also [Multimodal sentiment analysis](https://en.wikipedia.org/wiki/Multimodal_sentiment_analysis "Multimodal sentiment analysis")) Sentiment analysis involves identifying and classifying the emotional tone expressed in text. This technique involves analyzing text to determine whether the expressed sentiment is positive, negative, or neutral. Models for sentiment classification typically utilize inputs such as [word n-grams](https://en.wikipedia.org/wiki/Word_n-gram_language_model "Word n-gram language model"), [Term Frequency-Inverse Document Frequency](https://en.wikipedia.org/wiki/Term_frequency-inverse_document_frequency "Term frequency-inverse document frequency") (TF-IDF) features, hand-generated features, or employ [deep learning](https://en.wikipedia.org/wiki/Deep_learning "Deep learning") models designed to recognize both long-term and short-term dependencies in text sequences. The applications of sentiment analysis are diverse, extending to tasks such as categorizing customer reviews on various online platforms.[\[26\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-:0-26) [Terminology extraction](https://en.wikipedia.org/wiki/Terminology_extraction "Terminology extraction") The goal of terminology extraction is to automatically extract relevant terms from a given corpus. [Word-sense disambiguation](https://en.wikipedia.org/wiki/Word-sense_disambiguation "Word-sense disambiguation") (WSD) Many words have more than one [meaning](https://en.wikipedia.org/wiki/Meaning_\(linguistics\) "Meaning (linguistics)"); we have to select the meaning which makes the most sense in context. For this problem, we are typically given a list of words and associated word senses, e.g. from a dictionary or an online resource such as [WordNet](https://en.wikipedia.org/wiki/WordNet "WordNet"). [Entity linking](https://en.wikipedia.org/wiki/Entity_linking "Entity linking") Many words—typically proper names—refer to [named entities](https://en.wikipedia.org/wiki/Named_entity "Named entity"); here we have to select the entity (a famous individual, a location, a company, etc.) which is referred to in context. ### Relational semantics (semantics of individual sentences) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=12 "Edit section: Relational semantics (semantics of individual sentences)")\] [Relationship extraction](https://en.wikipedia.org/wiki/Relationship_extraction "Relationship extraction") Given a chunk of text, identify the relationships among named entities (e.g. who is married to whom). [Semantic parsing](https://en.wikipedia.org/wiki/Semantic_parsing "Semantic parsing") Given a piece of text (typically a sentence), produce a formal representation of its semantics, either as a graph (e.g., in [AMR parsing](https://en.wikipedia.org/wiki/Abstract_Meaning_Representation "Abstract Meaning Representation")) or in accordance with a logical formalism (e.g., in [DRT parsing](https://en.wikipedia.org/wiki/Discourse_representation_theory "Discourse representation theory")). This challenge typically includes aspects of several more elementary NLP tasks from semantics (e.g., semantic role labelling, word-sense disambiguation) and can be extended to include full-fledged discourse analysis (e.g., discourse analysis, coreference; see [Natural language understanding](https://en.wikipedia.org/wiki/Natural_language_processing#Natural_language_understanding) below). [Semantic role labelling](https://en.wikipedia.org/wiki/Semantic_role_labeling "Semantic role labeling") (see also implicit semantic role labelling below) Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https://en.wikipedia.org/wiki/Frame_semantics_\(linguistics\) "Frame semantics (linguistics)")), then identify and classify the frame elements ([semantic roles](https://en.wikipedia.org/wiki/Semantic_roles "Semantic roles")). ### Discourse (semantics beyond individual sentences) \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=13 "Edit section: Discourse (semantics beyond individual sentences)")\] [Coreference resolution](https://en.wikipedia.org/wiki/Coreference "Coreference") Given a sentence or larger chunk of text, determine which words ("mentions") refer to the same objects ("entities"). [Anaphora resolution](https://en.wikipedia.org/wiki/Anaphora_resolution "Anaphora resolution") is a specific example of this task, and is specifically concerned with matching up [pronouns](https://en.wikipedia.org/wiki/Pronoun "Pronoun") with the nouns or names to which they refer. The more general task of coreference resolution also includes identifying so-called "bridging relationships" involving [referring expressions](https://en.wikipedia.org/wiki/Referring_expression "Referring expression"). For example, in a sentence such as "He entered John's house through the front door", "the front door" is a referring expression and the bridging relationship to be identified is the fact that the door being referred to is the front door of John's house (rather than of some other structure that might also be referred to). [Discourse analysis](https://en.wikipedia.org/wiki/Discourse_analysis "Discourse analysis") This rubric includes several related tasks. One task is discourse parsing, i.e., identifying the [discourse](https://en.wikipedia.org/wiki/Discourse "Discourse") structure of a connected text, i.e. the nature of the discourse relationships between sentences (e.g. elaboration, explanation, contrast). Another possible task is recognizing and classifying the [speech acts](https://en.wikipedia.org/wiki/Speech_act "Speech act") in a chunk of text (e.g. yes–no question, content question, statement, assertion, etc.). Implicit semantic role labelling Given a single sentence, identify and disambiguate semantic predicates (e.g., verbal [frames](https://en.wikipedia.org/wiki/Frame_semantics_\(linguistics\) "Frame semantics (linguistics)")) and their explicit semantic roles in the current sentence (see [Semantic role labelling](https://en.wikipedia.org/wiki/Natural_language_processing#Semantic_role_labelling) above). Then, identify semantic roles that are not explicitly realized in the current sentence, classify them into arguments that are explicitly realized elsewhere in the text and those that are not specified, and resolve the former against the local text. A closely related task is zero anaphora resolution, i.e., the extension of coreference resolution to [pro-drop languages](https://en.wikipedia.org/wiki/Pro-drop_language "Pro-drop language"). [Recognizing textual entailment](https://en.wikipedia.org/wiki/Textual_entailment "Textual entailment") Given two text fragments, determine if one being true entails the other, entails the other's negation, or allows the other to be either true or false.[\[33\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-rte:11-33) [Topic segmentation](https://en.wikipedia.org/wiki/Topic_segmentation "Topic segmentation") and recognition Given a chunk of text, separate it into segments each of which is devoted to a topic, and identify the topic of the segment. [Argument mining](https://en.wikipedia.org/wiki/Argument_mining "Argument mining") The goal of argument mining is the automatic extraction and identification of argumentative structures from [natural language](https://en.wikipedia.org/wiki/Natural_language "Natural language") text with the aid of computer programs.[\[34\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-34) Such argumentative structures include the premise, conclusions, the [argument scheme](https://en.wikipedia.org/wiki/Argument_scheme "Argument scheme") and the relationship between the main and subsidiary argument, or the main and counter-argument within discourse.[\[35\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-35)[\[36\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-36) ### Higher-level NLP applications \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=14 "Edit section: Higher-level NLP applications")\] [![](https://upload.wikimedia.org/wikipedia/commons/thumb/4/43/Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png/250px-Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png)](https://en.wikipedia.org/wiki/File:Use_of_Firefox%27_%27Translate_selection_to_English%27_machine_translation_on_Commons_talk_meta_pages_02.png) Machine translation in Firefox [Automatic summarization](https://en.wikipedia.org/wiki/Automatic_summarization "Automatic summarization") (text summarization) Produce a readable summary of a chunk of text. Often used to provide summaries of the text of a known type, such as research papers, articles in the financial section of a newspaper. Grammatical error correction Grammatical error detection and correction involves a great band-width of problems on all levels of linguistic analysis (phonology/orthography, morphology, syntax, semantics, pragmatics). Grammatical error correction is impactful since it affects hundreds of millions of people that use or acquire English as a second language. It has thus been subject to a number of shared tasks since 2011.[\[37\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-37)[\[38\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-38)[\[39\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-39) As far as orthography, morphology, syntax and certain aspects of semantics are concerned, and due to the development of powerful neural language models such as [GPT-2](https://en.wikipedia.org/wiki/GPT-2 "GPT-2"), this can now (2019) be considered a largely solved problem and is being marketed in various commercial applications. [Logic translation](https://en.wikipedia.org/wiki/Logic_translation "Logic translation") Translate a text from a natural language into formal logic. [Machine translation](https://en.wikipedia.org/wiki/Machine_translation "Machine translation") (MT) Automatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed "[AI-complete](https://en.wikipedia.org/wiki/AI-complete "AI-complete")", i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) to solve properly. [Natural language understanding](https://en.wikipedia.org/wiki/Natural_language_understanding "Natural language understanding") (NLU) Convert chunks of text into more formal representations such as [first-order logic](https://en.wikipedia.org/wiki/First-order_logic "First-order logic") structures that are easier for [computer](https://en.wikipedia.org/wiki/Computer "Computer") programs to manipulate. Natural language understanding involves the identification of the intended semantic from the multiple possible semantics which can be derived from a natural language expression which usually takes the form of organized notations of natural language concepts. Introduction and creation of language metamodel and ontology are efficient however empirical solutions. An explicit formalization of natural language semantics without confusions with implicit assumptions such as [closed-world assumption](https://en.wikipedia.org/wiki/Closed-world_assumption "Closed-world assumption") (CWA) vs. [open-world assumption](https://en.wikipedia.org/wiki/Open-world_assumption "Open-world assumption"), or subjective Yes/No vs. objective True/False is expected for the construction of a basis of semantics formalization.[\[40\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-40) [Natural language generation](https://en.wikipedia.org/wiki/Natural_language_generation "Natural language generation") (NLG): Convert information from computer databases or semantic intents into readable human language. Book generation Not an NLP task proper but an extension of natural language generation and other NLP tasks is the creation of full-fledged books. The first machine-generated book was created by a rule-based system in 1984 (Racter, The policeman's beard is half-constructed).[\[41\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-41) The first published work by a neural network was published in 2018, [1 the Road](https://en.wikipedia.org/wiki/1_the_Road "1 the Road"), marketed as a novel, contains sixty million words. Both these systems are basically elaborate but non-sensical (semantics-free) [language models](https://en.wikipedia.org/wiki/Language_model "Language model"). The first machine-generated science book was published in 2019 (Beta Writer, Lithium-Ion Batteries, Springer, Cham).[\[42\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-42) Unlike Racter and 1 the Road, this is grounded on factual knowledge and based on text summarization. [Document AI](https://en.wikipedia.org/wiki/Document_AI "Document AI") A Document AI platform sits on top of the NLP technology enabling users with no prior experience of artificial intelligence, machine learning or NLP to quickly train a computer to extract the specific data they need from different document types. NLP-powered Document AI enables non-technical teams to quickly access information hidden in documents, for example, lawyers, business analysts and accountants.[\[43\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-43) [Dialogue management](https://en.wikipedia.org/wiki/Dialogue_system "Dialogue system") Computer systems intended to converse with a human. [Question answering](https://en.wikipedia.org/wiki/Question_answering "Question answering") Given a human-language question, determine its answer. Typical questions have a specific right answer (such as "What is the capital of Canada?"), but sometimes open-ended questions are also considered (such as "What is the meaning of life?"). [Text-to-image generation](https://en.wikipedia.org/wiki/Text-to-image_generation "Text-to-image generation") Given a description of an image, generate an image that matches the description.[\[44\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-44) Text-to-scene generation Given a description of a scene, generate a [3D model](https://en.wikipedia.org/wiki/3D_model "3D model") of the scene.[\[45\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-45)[\[46\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-46) [Text-to-video](https://en.wikipedia.org/wiki/Text-to-video_model "Text-to-video model") Given a description of a video, generate a video that matches the description.[\[47\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-47)[\[48\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-48) ## General tendencies and (possible) future directions \[[edit](https://en.wikipedia.org/w/index.php?title=Natural_language_processing&action=edit&section=15 "Edit section: General tendencies and (possible) future directions")\] Based on long-standing trends in the field, it is possible to extrapolate future directions of NLP. As of 2020, three trends among the topics of the long-standing series of CoNLL Shared Tasks can be observed:[\[49\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-49) - Interest on increasingly abstract, "cognitive" aspects of natural language (1999–2001: shallow parsing, 2002–03: named entity recognition, 2006–09/2017–18: dependency syntax, 2004–05/2008–09 semantic role labelling, 2011–12 coreference, 2015–16: discourse parsing, 2019: semantic parsing). - Increasing interest in multilinguality, and, potentially, multimodality (English since 1999; Spanish, Dutch since 2002; German since 2003; Bulgarian, Danish, Japanese, Portuguese, Slovenian, Swedish, Turkish since 2006; Basque, Catalan, Chinese, Greek, Hungarian, Italian, Turkish since 2007; Czech since 2009; Arabic since 2012; 2017: 40+ languages; 2018: 60+/100+ languages) - Elimination of symbolic representations (rule-based over supervised towards weakly supervised methods, representation learning and end-to-end systems) Most higher-level NLP applications involve aspects that emulate intelligent behavior and apparent comprehension of natural language. More broadly speaking, the technical operationalization of increasingly advanced aspects of cognitive behavior represents one of the developmental trajectories of NLP (see trends among CoNLL shared tasks above). [Cognition](https://en.wikipedia.org/wiki/Cognition "Cognition") refers to "the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses."[\[50\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-50) [Cognitive science](https://en.wikipedia.org/wiki/Cognitive_science "Cognitive science") is the interdisciplinary, scientific study of the mind and its processes.[\[51\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-51) [Cognitive linguistics](https://en.wikipedia.org/wiki/Cognitive_linguistics "Cognitive linguistics") is an interdisciplinary branch of linguistics, combining knowledge and research from both psychology and linguistics.[\[52\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-52) Especially during the age of [symbolic NLP](https://en.wikipedia.org/wiki/Natural_language_processing#Symbolic_NLP_\(1950s_%E2%80%93_early_1990s\)), the area of computational linguistics maintained strong ties with cognitive studies. As an example, [George Lakoff](https://en.wikipedia.org/wiki/George_Lakoff "George Lakoff") offers a methodology to build natural language processing (NLP) algorithms through the perspective of cognitive science, along with the findings of cognitive linguistics,[\[53\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-53) with two defining aspects: 1. Apply the theory of [conceptual metaphor](https://en.wikipedia.org/wiki/Conceptual_metaphor "Conceptual metaphor"), explained by Lakoff as "the understanding of one idea, in terms of another" which provides an idea of the intent of the author.[\[54\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-54) For example, consider the English word big. When used in a comparison ("That is a big tree"), the author's intent is to imply that the tree is physically large relative to other trees or the authors experience. When used metaphorically ("Tomorrow is a big day"), the author's intent to imply importance. The intent behind other usages, like in "She is a big person", will remain somewhat ambiguous to a person and a cognitive NLP algorithm alike without additional information. 2. Assign relative measures of meaning to a word, phrase, sentence or piece of text based on the information presented before and after the piece of text being analyzed, e.g., by means of a [probabilistic context-free grammar](https://en.wikipedia.org/wiki/Probabilistic_context-free_grammar "Probabilistic context-free grammar") (PCFG). The mathematical equation for such algorithms is presented in [US Patent 9269353](https://worldwide.espacenet.com/patent/search/family/055314712/publication/US9269353B1?q=pn%3DUS9269353):[\[55\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-55) ![{\\displaystyle {RMM(token\_{N})}={PMM(token\_{N})}\\times {\\frac {1}{2d}}\\left(\\sum \_{i=-d}^{d}{((PMM(token\_{N})}\\times {PF(token\_{N-i},token\_{N},token\_{N+i}))\_{i}}\\right)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/43ccadd794a4b84e20d1209997a463342e0dfbfe) Where RMM is the relative measure of meaning token is any block of text, sentence, phrase or word N is the number of tokens being analyzed PMM is the probable measure of meaning based on a corpora d is the non zero location of the token along the sequence of N tokens PF is the probability function specific to a language Ties with cognitive linguistics are part of the historical heritage of NLP, but they have been less frequently addressed since the statistical turn during the 1990s. Nevertheless, approaches to develop cognitive models towards technically operationalizable frameworks have been pursued in the context of various frameworks, e.g., of cognitive grammar,[\[56\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-56) functional grammar,[\[57\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-57) construction grammar,[\[58\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-58) computational psycholinguistics and cognitive neuroscience (e.g., [ACT-R](https://en.wikipedia.org/wiki/ACT-R "ACT-R")), however, with limited uptake in mainstream NLP (as measured by presence on major conferences[\[59\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-59) of the [ACL](https://en.wikipedia.org/wiki/Association_for_Computational_Linguistics "Association for Computational Linguistics")). More recently, ideas of cognitive NLP have been revived as an approach to achieve [explainability](https://en.wikipedia.org/wiki/Explainable_artificial_intelligence "Explainable artificial intelligence"), e.g., under the notion of "cognitive AI".[\[60\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-60) Likewise, ideas of cognitive NLP are inherent to neural models [multimodal](https://en.wikipedia.org/wiki/Multimodal_interaction "Multimodal interaction") NLP (although rarely made explicit)[\[61\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-61) and developments in [artificial intelligence](https://en.wikipedia.org/wiki/Artificial_intelligence "Artificial intelligence"), specifically tools and technologies using [large language model](https://en.wikipedia.org/wiki/Large_language_model "Large language model") approaches[\[62\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-62) and new directions in [artificial general intelligence](https://en.wikipedia.org/wiki/Artificial_general_intelligence "Artificial general intelligence") based on the [free energy principle](https://en.wikipedia.org/wiki/Free_energy_principle "Free energy principle")[\[63\]](https://en.wikipedia.org/wiki/Natural_language_processing#cite_note-63) by British neuroscientist and theoretician at University College London [Karl J. Friston](https://en.wikipedia.org/wiki/Karl_J._Friston "Karl J. Friston"). - [1 the Road](https://en.wikipedia.org/wiki/1_the_Road "1 the Road") - [Artificial intelligence detection software](https://en.wikipedia.org/wiki/Artificial_intelligence_detection_software "Artificial intelligence detection software") - [Automated essay scoring](https://en.wikipedia.org/wiki/Automated_essay_scoring "Automated essay scoring") - [Biomedical text mining](https://en.wikipedia.org/wiki/Biomedical_text_mining "Biomedical text mining") - [Compound term processing](https://en.wikipedia.org/wiki/Compound_term_processing "Compound term processing") - [Computational linguistics](https://en.wikipedia.org/wiki/Computational_linguistics "Computational linguistics") - [Computer-assisted reviewing](https://en.wikipedia.org/wiki/Computer-assisted_reviewing "Computer-assisted reviewing") - [Controlled natural language](https://en.wikipedia.org/wiki/Controlled_natural_language "Controlled natural language") - [Deep learning](https://en.wikipedia.org/wiki/Deep_learning "Deep learning") - [Deep linguistic processing](https://en.wikipedia.org/wiki/Deep_linguistic_processing "Deep linguistic processing") - [Distributional semantics](https://en.wikipedia.org/wiki/Distributional_semantics "Distributional semantics") - [Foreign language reading aid](https://en.wikipedia.org/wiki/Foreign_language_reading_aid "Foreign language reading aid") - [Foreign language writing aid](https://en.wikipedia.org/wiki/Foreign_language_writing_aid "Foreign language writing aid") - [Information extraction](https://en.wikipedia.org/wiki/Information_extraction "Information extraction") - [Information retrieval](https://en.wikipedia.org/wiki/Information_retrieval "Information retrieval") - [Language and Communication Technologies](https://en.wikipedia.org/wiki/Language_and_Communication_Technologies "Language and Communication Technologies") - [Language model](https://en.wikipedia.org/wiki/Language_model "Language model") - [Language technology](https://en.wikipedia.org/wiki/Language_technology "Language technology") - [Latent semantic indexing](https://en.wikipedia.org/wiki/Latent_semantic_indexing "Latent semantic indexing") - [Multi-agent system](https://en.wikipedia.org/wiki/Multi-agent_system "Multi-agent system") - [Native-language identification](https://en.wikipedia.org/wiki/Native-language_identification "Native-language identification") - [Natural-language programming](https://en.wikipedia.org/wiki/Natural-language_programming "Natural-language programming") - [Natural-language understanding](https://en.wikipedia.org/wiki/Natural-language_understanding "Natural-language understanding") - [Natural-language search](https://en.wikipedia.org/wiki/Natural-language_user_interface "Natural-language user interface") - [Outline of natural language processing](https://en.wikipedia.org/wiki/Outline_of_natural_language_processing "Outline of natural language processing") - [Query expansion](https://en.wikipedia.org/wiki/Query_expansion "Query expansion") - [Query understanding](https://en.wikipedia.org/wiki/Query_understanding "Query understanding") - [Reification (linguistics)](https://en.wikipedia.org/wiki/Reification_\(linguistics\) "Reification (linguistics)") - [Speech processing](https://en.wikipedia.org/wiki/Speech_processing "Speech processing") - [Spoken dialogue systems](https://en.wikipedia.org/wiki/Spoken_dialogue_systems "Spoken dialogue systems") - [Text-proofing](https://en.wikipedia.org/wiki/Text-proofing "Text-proofing") - [Text simplification](https://en.wikipedia.org/wiki/Text_simplification "Text simplification") - [Transformer (machine learning model)](https://en.wikipedia.org/wiki/Transformer_\(machine_learning_model\) "Transformer (machine learning model)") - [Truecasing](https://en.wikipedia.org/wiki/Truecasing "Truecasing") - [Question answering](https://en.wikipedia.org/wiki/Question_answering "Question answering") - [Word2vec](https://en.wikipedia.org/wiki/Word2vec "Word2vec") 1. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-nlpintro_1-0) Eisenstein, Jacob (October 1, 2019). [Introduction to Natural Language Processing](https://mitpress.mit.edu/9780262042840/introduction-to-natural-language-processing/). The MIT Press. p. 1. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-04284-0](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-04284-0 "Special:BookSources/978-0-262-04284-0") . 2. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-2) ["NLP"](https://cs.stanford.edu/people/eroberts/courses/soco/projects/2004-05/nlp/overview_history.html). 3. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-3) Hutchins, J. (2005). ["The history of machine translation in a nutshell"](https://web.archive.org/web/20190713103044/http://www.hutchinsweb.me.uk/Nutshell-2005.pdf) (PDF). Archived from [the original](http://www.hutchinsweb.me.uk/Nutshell-2005.pdf) (PDF) on 2019-07-13. Retrieved 2019-02-04. \[[self-published source](https://en.wikipedia.org/wiki/Wikipedia:Verifiability#Self-published_sources "Wikipedia:Verifiability")\] 4. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-4) "ALPAC: the (in)famous report", John Hutchins, MT News International, no. 14, June 1996, pp. 9–12. 5. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-5) [Crevier 1993](https://en.wikipedia.org/wiki/Natural_language_processing#CITEREFCrevier1993), pp. 146–148, see also [Buchanan 2005](https://en.wikipedia.org/wiki/Natural_language_processing#CITEREFBuchanan2005), p. 56: "Early programs were necessarily limited in scope by the size and speed of memory" 6. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-6) [Koskenniemi, Kimmo](https://en.wikipedia.org/wiki/Kimmo_Koskenniemi "Kimmo Koskenniemi") (1983), [Two-level morphology: A general computational model of word-form recognition and production](https://web.archive.org/web/20181221032913/http://www.ling.helsinki.fi/~koskenni/doc/Two-LevelMorphology.pdf) (PDF), Department of General Linguistics, [University of Helsinki](https://en.wikipedia.org/wiki/University_of_Helsinki "University of Helsinki"), archived from [the original](http://www.ling.helsinki.fi/~koskenni/doc/Two-LevelMorphology.pdf) (PDF) on 2018-12-21, retrieved 2020-08-20 7. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-7) Joshi, A. K., & Weinstein, S. (1981, August). [Control of Inference: Role of Some Aspects of Discourse Structure-Centering](https://www.ijcai.org/Proceedings/81-1/Papers/071.pdf). In IJCAI (pp. 385–387). 8. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-8) Guida, G.; Mauri, G. (July 1986). "Evaluation of natural language processing systems: Issues and approaches". Proceedings of the IEEE. 74 (7): 1026–1035\. [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[1986IEEEP..74.1026G](https://ui.adsabs.harvard.edu/abs/1986IEEEP..74.1026G). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1109/PROC.1986.13580](https://doi.org/10.1109%2FPROC.1986.13580). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1558-2256](https://search.worldcat.org/issn/1558-2256). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [30688575](https://api.semanticscholar.org/CorpusID:30688575). 9. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-9) Chomskyan linguistics encourages the investigation of "[corner cases](https://en.wikipedia.org/wiki/Corner_case "Corner case")" that stress the limits of its theoretical models (comparable to [pathological](https://en.wikipedia.org/wiki/Pathological_\(mathematics\) "Pathological (mathematics)") phenomena in mathematics), typically created using [thought experiments](https://en.wikipedia.org/wiki/Thought_experiment "Thought experiment"), rather than the systematic investigation of typical phenomena that occur in real-world data, as is the case in [corpus linguistics](https://en.wikipedia.org/wiki/Corpus_linguistics "Corpus linguistics"). The creation and use of such [corpora](https://en.wikipedia.org/wiki/Text_corpus "Text corpus") of real-world data is a fundamental part of machine-learning algorithms for natural language processing. In addition, theoretical underpinnings of Chomskyan linguistics such as the so-called "[poverty of the stimulus](https://en.wikipedia.org/wiki/Poverty_of_the_stimulus "Poverty of the stimulus")" argument entail that general learning algorithms, as are typically used in machine learning, cannot be successful in language processing. As a result, the Chomskyan paradigm discouraged the application of such models to language processing. 10. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-10) Bengio, Yoshua; Ducharme, Réjean; Vincent, Pascal; Janvin, Christian (March 1, 2003). ["A neural probabilistic language model"](https://dl.acm.org/doi/10.5555/944919.944966). The Journal of Machine Learning Research. 3: 1137–1155 – via ACM Digital Library. 11. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-11) Mikolov, Tomáš; Karafiát, Martin; Burget, Lukáš; Černocký, Jan; Khudanpur, Sanjeev (26 September 2010). ["Recurrent neural network based language model"](https://gwern.net/doc/ai/nn/rnn/2010-mikolov.pdf) (PDF). Interspeech 2010. pp. 1045–1048\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.21437/Interspeech.2010-343](https://doi.org/10.21437%2FInterspeech.2010-343). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [17048224](https://api.semanticscholar.org/CorpusID:17048224). 12. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-goldberg:nnlp17_12-0) Goldberg, Yoav (2016). ["A Primer on Neural Network Models for Natural Language Processing"](https://doi.org/10.1613%2Fjair.4992). Journal of Artificial Intelligence Research. 57: 345–420\. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1807\.10854](https://arxiv.org/abs/1807.10854). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1613/jair.4992](https://doi.org/10.1613%2Fjair.4992). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [8273530](https://api.semanticscholar.org/CorpusID:8273530). 13. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-goodfellow:book16_13-0) Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2016). [Deep Learning](http://www.deeplearningbook.org/). MIT Press. 14. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-jozefowicz:lm16_14-0) Jozefowicz, Rafal; Vinyals, Oriol; Schuster, Mike; Shazeer, Noam; Wu, Yonghui (2016). Exploring the Limits of Language Modeling. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1602\.02410](https://arxiv.org/abs/1602.02410). [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[2016arXiv160202410J](https://ui.adsabs.harvard.edu/abs/2016arXiv160202410J). 15. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-choe:emnlp16_15-0) Choe, Do Kook; Charniak, Eugene. ["Parsing as Language Modeling"](https://web.archive.org/web/20181023034804/https://aclanthology.coli.uni-saarland.de/papers/D16-1257/d16-1257). Emnlp 2016. Archived from [the original](https://aclanthology.coli.uni-saarland.de/papers/D16-1257/d16-1257) on 2018-10-23. Retrieved 2018-10-22. 16. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-vinyals:nips15_16-0) Vinyals, Oriol; et al. (2014). ["Grammar as a Foreign Language"](https://papers.nips.cc/paper/5635-grammar-as-a-foreign-language.pdf) (PDF). Nips2015. [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[1412\.7449](https://arxiv.org/abs/1412.7449). [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[2014arXiv1412.7449V](https://ui.adsabs.harvard.edu/abs/2014arXiv1412.7449V). 17. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-17) Turchin, Alexander; Florez Builes, Luisa F. (2021-03-19). ["Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120048). Journal of Diabetes Science and Technology. 15 (3): 553–560\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1177/19322968211000831](https://doi.org/10.1177%2F19322968211000831). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1932-2968](https://search.worldcat.org/issn/1932-2968). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [8120048](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120048). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [33736486](https://pubmed.ncbi.nlm.nih.gov/33736486). 18. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-18) Lee, Jennifer; Yang, Samuel; Holland-Hall, Cynthia; Sezgin, Emre; Gill, Manjot; Linwood, Simon; Huang, Yungui; Hoffman, Jeffrey (2022-06-10). ["Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9233261). JMIR Medical Informatics. 10 (6) e38482. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.2196/38482](https://doi.org/10.2196%2F38482). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [2291-9694](https://search.worldcat.org/issn/2291-9694). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [9233261](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9233261). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [35687381](https://pubmed.ncbi.nlm.nih.gov/35687381). 19. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-winograd:shrdlu71_19-0) Winograd, Terry (1971). [Procedures as a Representation for Data in a Computer Program for Understanding Natural Language](http://hci.stanford.edu/winograd/shrdlu/) (Thesis). 20. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-schank77_20-0) Schank, Roger C.; Abelson, Robert P. (1977). Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures. Hillsdale: Erlbaum. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [0-470-99033-3](https://en.wikipedia.org/wiki/Special:BookSources/0-470-99033-3 "Special:BookSources/0-470-99033-3") . 21. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-johnson:eacl:ilcl09_21-0) [Mark Johnson. How the statistical revolution changes (computational) linguistics.](http://www.aclweb.org/anthology/W09-0103) Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics. 22. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-resnik:langlog11_22-0) [Philip Resnik. Four revolutions.](http://languagelog.ldc.upenn.edu/nll/?p=2946) Language Log, February 5, 2011. 23. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-23) Socher, Richard. ["Deep Learning For NLP-ACL 2012 Tutorial"](https://web.archive.org/web/20210414054126/https://www.socher.org/index.php/Main/DeepLearningForNLP-ACL2012Tutorial). www.socher.org. Archived from [the original](https://www.socher.org/index.php/Main/DeepLearningForNLP-ACL2012Tutorial) on 2021-04-14. Retrieved 2020-08-17. This was an early Deep Learning tutorial at the ACL 2012 and met with both interest and (at the time) skepticism by most participants. Until then, neural learning was basically rejected because of its lack of statistical interpretability. Until 2015, deep learning had evolved into the major framework of NLP. \[Link is broken, try <http://web.stanford.edu/class/cs224n/>\] 24. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-24) Segev, Elad (2022). [Semantic Network Analysis in Social Sciences](https://www.routledge.com/Semantic-Network-Analysis-in-Social-Sciences/Segev/p/book/9780367636524). London: Routledge. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-367-63652-4](https://en.wikipedia.org/wiki/Special:BookSources/978-0-367-63652-4 "Special:BookSources/978-0-367-63652-4") . [Archived](https://web.archive.org/web/20211205140726/https://www.routledge.com/Semantic-Network-Analysis-in-Social-Sciences/Segev/p/book/9780367636524) from the original on 5 December 2021. Retrieved 5 December 2021. 25. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-25) Yi, Chucai; [Tian, Yingli](https://en.wikipedia.org/wiki/Yingli_Tian "Yingli Tian") (2012), "Assistive Text Reading from Complex Background for Blind Persons", Camera-Based Document Analysis and Recognition, Lecture Notes in Computer Science, vol. 7139, Springer Berlin Heidelberg, pp. 15–28, [CiteSeerX](https://en.wikipedia.org/wiki/CiteSeerX_\(identifier\) "CiteSeerX (identifier)") [10\.1.1.668.869](https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.668.869), [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1007/978-3-642-29364-1\_2](https://doi.org/10.1007%2F978-3-642-29364-1_2), [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-3-642-29363-4](https://en.wikipedia.org/wiki/Special:BookSources/978-3-642-29363-4 "Special:BookSources/978-3-642-29363-4") `{{citation}}`: CS1 maint: work parameter with ISBN ([link](https://en.wikipedia.org/wiki/Category:CS1_maint:_work_parameter_with_ISBN "Category:CS1 maint: work parameter with ISBN")) 26. ^ [*a](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-:0_26-0) [b](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-:0_26-1) ["Natural Language Processing (NLP) - A Complete Guide"](https://www.deeplearning.ai/resources/natural-language-processing/). www.deeplearning.ai. 2023-01-11. Retrieved 2024-05-05. 27. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-27)* ["GeeksforGeeks. (n.d.). Tokenization in natural language processing (NLP). GeeksforGeeks"](https://www.geeksforgeeks.org/nlp/tokenization-in-natural-language-processing-nlp/). Geeksforgeeks. 25 June 2024. 28. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-28) Torabi, Mohhamadreza; Ghasemi, Fahimeh. ["Integrating data augmentation and BERT-based deep learning for predicting alpha-glucosidase inhibitors derived from Black Cohosh"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12391535). Scientific Reports. 1 (1): 1–15\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1038/s41598-025-14699-1](https://doi.org/10.1038%2Fs41598-025-14699-1). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [12391535](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12391535). 29. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-29) ["What is Natural Language Processing? Intro to NLP in Machine Learning"](https://www.gyansetu.in/what-is-natural-language-processing/). GyanSetu\!. 2020-12-06. Retrieved 2021-01-09. 30. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-30) Kishorjit, N.; Vidya, Raj RK.; Nirmal, Y.; Sivaji, B. (2012). ["Manipuri Morpheme Identification"](http://aclweb.org/anthology//W/W12/W12-5008.pdf) (PDF). Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing (SANLP). COLING 2012, Mumbai, December 2012: 95–108\. `{{cite journal}}`: CS1 maint: location ([link](https://en.wikipedia.org/wiki/Category:CS1_maint:_location "Category:CS1 maint: location")) 31. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-31) Klein, Dan; Manning, Christopher D. (2002). ["Natural language grammar induction using a constituent-context model"](http://papers.nips.cc/paper/1945-natural-language-grammar-induction-using-a-constituent-context-model.pdf) (PDF). Advances in Neural Information Processing Systems. 32. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-32) Kariampuzha, William; Alyea, Gioconda; Qu, Sue; Sanjak, Jaleal; Mathé, Ewy; Sid, Eric; Chatelaine, Haley; Yadaw, Arjun; Xu, Yanji; Zhu, Qian (2023). ["Precision information extraction for rare disease epidemiology at scale"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9972634). Journal of Translational Medicine. 21 (1): 157. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1186/s12967-023-04011-y](https://doi.org/10.1186%2Fs12967-023-04011-y). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [9972634](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9972634). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [36855134](https://pubmed.ncbi.nlm.nih.gov/36855134). 33. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-rte:11_33-0) PASCAL Recognizing Textual Entailment Challenge (RTE-7) <https://tac.nist.gov//2011/RTE/> 34. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-34) Lippi, Marco; Torroni, Paolo (2016-04-20). ["Argumentation Mining: State of the Art and Emerging Trends"](https://dl.acm.org/doi/10.1145/2850417). ACM Transactions on Internet Technology. 16 (2): 1–25\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1145/2850417](https://doi.org/10.1145%2F2850417). [hdl](https://en.wikipedia.org/wiki/Hdl_\(identifier\) "Hdl (identifier)"):[11585/523460](https://hdl.handle.net/11585%2F523460). [ISSN](https://en.wikipedia.org/wiki/ISSN_\(identifier\) "ISSN (identifier)") [1533-5399](https://search.worldcat.org/issn/1533-5399). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [9561587](https://api.semanticscholar.org/CorpusID:9561587). 35. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-35) ["Argument Mining – IJCAI2016 Tutorial"](https://web.archive.org/web/20210418083659/https://www.i3s.unice.fr/~villata/tutorialIJCAI2016.html). www.i3s.unice.fr. Archived from [the original](https://www.i3s.unice.fr/~villata/tutorialIJCAI2016.html) on 2021-04-18. Retrieved 2021-03-09. 36. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-36) ["NLP Approaches to Computational Argumentation – ACL 2016, Berlin"](http://acl2016tutorial.arg.tech/). Retrieved 2021-03-09. 37. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-37) Administration. ["Centre for Language Technology (CLT)"](https://www.mq.edu.au/research/research-centres-groups-and-facilities/innovative-technologies/centres/centre-for-language-technology-clt). Macquarie University. Retrieved 2021-01-11. 38. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-38) ["Shared Task: Grammatical Error Correction"](https://www.comp.nus.edu.sg/~nlp/conll13st.html). www.comp.nus.edu.sg. Retrieved 2021-01-11. 39. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-39) ["Shared Task: Grammatical Error Correction"](https://www.comp.nus.edu.sg/~nlp/conll14st.html). www.comp.nus.edu.sg. Retrieved 2021-01-11. 40. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-40) Duan, Yucong; Cruz, Christophe (2011). ["Formalizing Semantic of Natural Language through Conceptualization from Existence"](https://web.archive.org/web/20111009135952/http://www.ijimt.org/abstract/100-E00187.htm). International Journal of Innovation, Management and Technology. 2 (1): 37–42\. Archived from [the original](http://www.ijimt.org/abstract/100-E00187.htm) on 2011-10-09. 41. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-41) ["U B U W E B :: Racter"](http://www.ubu.com/historical/racter/index.html). www.ubu.com. Retrieved 2020-08-17. 42. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-42) Writer, Beta (2019). Lithium-Ion Batteries. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1007/978-3-030-16800-1](https://doi.org/10.1007%2F978-3-030-16800-1). [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-3-030-16799-8](https://en.wikipedia.org/wiki/Special:BookSources/978-3-030-16799-8 "Special:BookSources/978-3-030-16799-8") . [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [155818532](https://api.semanticscholar.org/CorpusID:155818532). 43. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-43) ["Document Understanding AI on Google Cloud (Cloud Next '19) – YouTube"](https://ghostarchive.org/varchive/youtube/20211030/7dtl650D0y0). www.youtube.com. 11 April 2019. Archived from [the original](https://www.youtube.com/watch?v=7dtl650D0y0) on 2021-10-30. Retrieved 2021-01-11. 44. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-44) Robertson, Adi (2022-04-06). ["OpenAI's DALL-E AI image generator can now edit pictures, too"](https://www.theverge.com/2022/4/6/23012123/openai-clip-dalle-2-ai-text-to-image-generator-testing). The Verge. Retrieved 2022-06-07. 45. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-45) ["The Stanford Natural Language Processing Group"](https://nlp.stanford.edu/projects/text2scene.shtml). nlp.stanford.edu. Retrieved 2022-06-07. 46. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-46) Coyne, Bob; Sproat, Richard (2001-08-01). ["WordsEye"](https://doi.org/10.1145/383259.383316). Proceedings of the 28th annual conference on Computer graphics and interactive techniques. SIGGRAPH '01. New York, NY, USA: Association for Computing Machinery. pp. 487–496\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1145/383259.383316](https://doi.org/10.1145%2F383259.383316). [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1-58113-374-5](https://en.wikipedia.org/wiki/Special:BookSources/978-1-58113-374-5 "Special:BookSources/978-1-58113-374-5") . [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [3842372](https://api.semanticscholar.org/CorpusID:3842372). 47. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-47) ["Google announces AI advances in text-to-video, language translation, more"](https://venturebeat.com/ai/google-announces-ai-advances-in-text-to-video-language-translation-more/). VentureBeat. 2022-11-02. Retrieved 2022-11-09. 48. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-48) Vincent, James (2022-09-29). ["Meta's new text-to-video AI generator is like DALL-E for video"](https://www.theverge.com/2022/9/29/23378210/meta-text-to-video-ai-generation-make-a-video-model-dall-e). The Verge. Retrieved 2022-11-09. 49. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-49) ["Previous shared tasks \\| CoNLL"](https://www.conll.org/previous-tasks). www.conll.org. Retrieved 2021-01-11. 50. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-50) ["Cognition"](https://web.archive.org/web/20200715113427/https://www.lexico.com/definition/cognition). Lexico. [Oxford University Press](https://en.wikipedia.org/wiki/Oxford_University_Press "Oxford University Press") and [Dictionary.com](https://en.wikipedia.org/wiki/Dictionary.com "Dictionary.com"). Archived from [the original](https://www.lexico.com/definition/cognition) on July 15, 2020. Retrieved 6 May 2020. 51. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-51) ["Ask the Cognitive Scientist"](http://www.aft.org/newspubs/periodicals/ae/summer2002/willingham.cfm). American Federation of Teachers. 8 August 2014. "Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind." 52. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-52) Robinson, Peter (2008). Handbook of Cognitive Linguistics and Second Language Acquisition. Routledge. pp. 3–8\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-805-85352-0](https://en.wikipedia.org/wiki/Special:BookSources/978-0-805-85352-0 "Special:BookSources/978-0-805-85352-0") . 53. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-53) Lakoff, George (1999). Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Philosophy; Appendix: The Neural Theory of Language Paradigm. New York Basic Books. pp. 569–583\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-465-05674-3](https://en.wikipedia.org/wiki/Special:BookSources/978-0-465-05674-3 "Special:BookSources/978-0-465-05674-3") . 54. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-54) Strauss, Claudia (1999). A Cognitive Theory of Cultural Meaning. Cambridge University Press. pp. 156–164\. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-521-59541-4](https://en.wikipedia.org/wiki/Special:BookSources/978-0-521-59541-4 "Special:BookSources/978-0-521-59541-4") . 55. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-55) [US patent 9269353](https://worldwide.espacenet.com/textdoc?DB=EPODOC&IDX=US9269353) 56. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-56) ["Universal Conceptual Cognitive Annotation (UCCA)"](https://universalconceptualcognitiveannotation.github.io/). Universal Conceptual Cognitive Annotation (UCCA). Retrieved 2021-01-11. 57. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-57) Rodríguez, F. C., & Mairal-Usón, R. (2016). [Building an RRG computational grammar](https://www.redalyc.org/pdf/1345/134549291020.pdf). Onomazein, (34), 86–117. 58. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-58) ["Fluid Construction Grammar – A fully operational processing system for construction grammars"](https://www.fcg-net.org/). Retrieved 2021-01-11. 59. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-59) ["ACL Member Portal \\| The Association for Computational Linguistics Member Portal"](https://www.aclweb.org/portal/). www.aclweb.org. Retrieved 2021-01-11. 60. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-60) ["Chunks and Rules"](https://www.w3.org/Data/demos/chunks/chunks.html). W3C. Retrieved 2021-01-11. 61. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-61) Socher, Richard; Karpathy, Andrej; Le, Quoc V.; Manning, Christopher D.; Ng, Andrew Y. (2014). ["Grounded Compositional Semantics for Finding and Describing Images with Sentences"](https://doi.org/10.1162%2Ftacl_a_00177). Transactions of the Association for Computational Linguistics. 2: 207–218\. [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1162/tacl\_a\_00177](https://doi.org/10.1162%2Ftacl_a_00177). [S2CID](https://en.wikipedia.org/wiki/S2CID_\(identifier\) "S2CID (identifier)") [2317858](https://api.semanticscholar.org/CorpusID:2317858). 62. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-62) Dasgupta, Ishita; Lampinen, Andrew K.; Chan, Stephanie C. Y.; Creswell, Antonia; Kumaran, Dharshan; McClelland, James L.; Hill, Felix (2022). "Language models show human-like content effects on reasoning, Dasgupta, Lampinen et al". [arXiv](https://en.wikipedia.org/wiki/ArXiv_\(identifier\) "ArXiv (identifier)"):[2207\.07051](https://arxiv.org/abs/2207.07051) \[[cs.CL](https://arxiv.org/archive/cs.CL)\]. 63. [^](https://en.wikipedia.org/wiki/Natural_language_processing#cite_ref-63) Friston, Karl J. (2022). Active Inference: The Free Energy Principle in Mind, Brain, and Behavior; Chapter 4 The Generative Models of Active Inference. The MIT Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-36997-8](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-36997-8 "Special:BookSources/978-0-262-36997-8") . - Bates, M (1995). ["Models of natural language understanding"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC40721). Proceedings of the National Academy of Sciences of the United States of America. 92 (22): 9977–9982\. [Bibcode](https://en.wikipedia.org/wiki/Bibcode_\(identifier\) "Bibcode (identifier)"):[1995PNAS...92.9977B](https://ui.adsabs.harvard.edu/abs/1995PNAS...92.9977B). [doi](https://en.wikipedia.org/wiki/Doi_\(identifier\) "Doi (identifier)"):[10\.1073/pnas.92.22.9977](https://doi.org/10.1073%2Fpnas.92.22.9977). [PMC](https://en.wikipedia.org/wiki/PMC_\(identifier\) "PMC (identifier)") [40721](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC40721). [PMID](https://en.wikipedia.org/wiki/PMID_\(identifier\) "PMID (identifier)") [7479812](https://pubmed.ncbi.nlm.nih.gov/7479812). - Steven Bird, Ewan Klein, and Edward Loper (2009). Natural Language Processing with Python. O'Reilly Media. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-596-51649-9](https://en.wikipedia.org/wiki/Special:BookSources/978-0-596-51649-9 "Special:BookSources/978-0-596-51649-9") . - [Kenna Hughes-Castleberry](https://en.wikipedia.org/w/index.php?title=Kenna_Hughes-Castleberry&action=edit&redlink=1 "Kenna Hughes-Castleberry (page does not exist)"), "A Murder Mystery Puzzle: The literary puzzle [Cain's Jawbone](https://en.wikipedia.org/wiki/Cain%27s_Jawbone "Cain's Jawbone"), which has stumped humans for decades, reveals the limitations of natural-language-processing algorithms", [Scientific American](https://en.wikipedia.org/wiki/Scientific_American "Scientific American"), vol. 329, no. 4 (November 2023), pp. 81–82. "This murder mystery competition has revealed that although NLP ([natural-language processing](https://en.wikipedia.org/wiki/Natural-language_processing "Natural-language processing")) models are capable of incredible feats, their abilities are very much limited by the amount of [context](https://en.wikipedia.org/wiki/Context_\(linguistics\) "Context (linguistics)") they receive. This \[...\] could cause \[difficulties\] for researchers who hope to use them to do things such as analyze [ancient languages](https://en.wikipedia.org/wiki/Ancient_language "Ancient language"). In some cases, there are few historical records on long-gone [civilizations](https://en.wikipedia.org/wiki/Civilization "Civilization") to serve as [training data](https://en.wikipedia.org/wiki/Training_data "Training data") for such a purpose." (p. 82.) - Daniel Jurafsky and James H. Martin (2008). Speech and Language Processing, 2nd edition. Pearson Prentice Hall. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-13-187321-6](https://en.wikipedia.org/wiki/Special:BookSources/978-0-13-187321-6 "Special:BookSources/978-0-13-187321-6") . - Mohamed Zakaria Kurdi (2016). Natural Language Processing and Computational Linguistics: speech, morphology, and syntax, Volume 1. ISTE-Wiley. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1848218482](https://en.wikipedia.org/wiki/Special:BookSources/978-1848218482 "Special:BookSources/978-1848218482") . - Mohamed Zakaria Kurdi (2017). Natural Language Processing and Computational Linguistics: semantics, discourse, and applications, Volume 2. ISTE-Wiley. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-1848219212](https://en.wikipedia.org/wiki/Special:BookSources/978-1848219212 "Special:BookSources/978-1848219212") . - Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze (2008). Introduction to Information Retrieval. Cambridge University Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-521-86571-5](https://en.wikipedia.org/wiki/Special:BookSources/978-0-521-86571-5 "Special:BookSources/978-0-521-86571-5") . [Official html and pdf versions available without charge.](http://nlp.stanford.edu/IR-book/) - Christopher D. Manning and Hinrich Schütze (1999). Foundations of Statistical Natural Language Processing. The MIT Press. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-262-13360-9](https://en.wikipedia.org/wiki/Special:BookSources/978-0-262-13360-9 "Special:BookSources/978-0-262-13360-9") . - David M. W. Powers and Christopher C. R. Turk (1989). Machine Learning of Natural Language. Springer-Verlag. [ISBN](https://en.wikipedia.org/wiki/ISBN_\(identifier\) "ISBN (identifier)") [978-0-387-19557-5](https://en.wikipedia.org/wiki/Special:BookSources/978-0-387-19557-5 "Special:BookSources/978-0-387-19557-5") . - [![Wikimedia Commons logo](https://upload.wikimedia.org/wikipedia/en/thumb/4/4a/Commons-logo.svg/20px-Commons-logo.svg.png)](https://en.wikipedia.org/wiki/File:Commons-logo.svg) Media related to [Natural language processing](https://commons.wikimedia.org/wiki/Category:Natural_language_processing "commons:Category:Natural language processing") at Wikimedia Commons
Shard	152 (laksa)
Root Hash	17790707453426894952
Unparsed URL	org,wikipedia!en,/wiki/Natural_language_processing s443