computational linguistics


A multidisciplinary field concerned with the processing of language by computers.

On its website, the Association for Computational Linguistics defines computational linguistics (CL) as "the scientific study of language from a computational perspective." The entry goes on to report that "the work of computational linguists is incorporated into many working systems today, including speech recognition systems, text-to-speech synthesizers, automated voice response systems, web search engines, text editors, language instruction materials, to name just a few."

Among the numerous subfields of computational linguistics is computational semantics, which deals with "linguistic meaning within a computational approach to natural language" (Oxford Handbook of Computational Linguistics, 2003). See Examples and Observations, below.

See also:

Examples and Observations:

  • "The field of computational linguistics (CL), together with its engineering domain of natural language processing (NLP), has exploded in recent years. It has developed rapidly from a relatively obscure adjunct of both AI [artificial intelligence] and formal linguistics into a thriving scientific discipline. It has also become an important area of industrial development. The focus of research in CL and NLP has shifted over the past three decades from the study of small prototypes and theoretical models to robust learning and processing systems applied to large corpora."
    (Introduction to The Handbook of Computational Linguistics and Natural Language Processing, ed. by Alexander Clark, Chris Fox, and Shalom Lappin. Wiley-Blackwell, 2013)
  • Computational Linguistics and Natural Language Processing
    "The two terms, Computational Linguistics (CL) and Natural Language Processing (NLP), have often been used interchangeably. However, these two terms represent two different streams of research which emphasize different aspects of our field. For example, while research on grammar and its formalisms in CL and research on parsing in NLP are closely related, their objectives are quite different. On one hand, researchers in CL have focused on revealing how surface strings of words systematically correspond to their meanings (in a compositional way) and have been interested in developing formalisms by which the correspondences are described. On the other hand, those in NLP are interested in more practical engineering issues involved in processing natural languages by computer, such as efficient algorithms for a program (parser) which computes the structure and/or the meaning of a given sentence."
    (Jun'ichi Tsujii, "Computational Linguistics and Natural Language Processing." Computational Linguistics and Intelligent Text Processing: 12th International Conference, ed. by Alexander Gelbukh. Springer, 2011)
  • A Multidisciplinary Field
    "Computational Linguistics has a long history, dating back to the Fifties, during which it developed a whole set of computational models and implementations, theories, methodologies and applications. . . . Since its origins, Computational Linguistics has been in an intermediate position between Computer Science and Artificial Intelligence, Linguistics and Cognitive Science, and Engineering. Computer Science itself shares its roots with Computational Linguistics: parsing, which is central for the design of compilers for programming languages (Aho & Ullmann 1977:6), is also the building block of any natural language processing engine, and both are the realizations of the Chomskian theory of formal languages (Chomsky 1957). . . .

    "By now, we can say that while Computational Linguists were, and are, more interested in the correctness and plausibility of their models, Engineers were, and are, more interested in the usability of tools and techniques, even if this entails the risk of 'dirty' solutions. The history of Computational Linguistics in the last few decades is much the history of the evolving relations between all these conjuring approaches."
    (Giacomo Ferrari, "State of the Art in Computational Linguistics." Linguistics Today: Facing a Greater Challenge, ed. by Piet van Sterkenburg. John Benjamins, 2004)
  • Computational Semantics: Meaning and Syntax
    "A main focus of computational semantics is to model the way in which the meanings of phrases and sentences are computed systematically from the meanings of their syntactic constituents. It is possible to factor this study into two main questions. The first is, how should meanings be represented? The second is, how are the semantic representations of syntactically complex expressions assembled from the meanings of their component parts? The second question implies that a system for semantic interpretation will work in tandem with a procedure for representing and parsing the syntactic structure of an input string. Given the assumption that generating a semantic interpretation for a sentence involves constructing a representation on the basis of its syntactic structure, a central issue for any semantic theory is the nature of the interface between syntax and semantics. A second important problem is to determine the precise nature of the general condition which is imposed by the requirement that a computational semantics define a systematic relation between the meaning of a sentence and the meanings of its constituents. A third major concern that a theory of semantics must deal with is the role of contextual and discourse factors in the interpretation of a sentence."
    (Shalom Lappin, "Semantics." The Oxford Handbook of Computational Linguistics, ed. by Ruslan Mitkov. Oxford University Press, 2003)

    "[In order to do computational semantics properly you need proficiency in at least three things: (i) proficiency in computation (you need to be proficient in the use of at least one suitable programming language), (ii) proficiency in syntax, and (iii) proficiency in semantics . . ..

    "[C]omputing semantic representations from 'raw data' isn't all there is to computational semantics. Computing semantic representations wouldn't be of much use to us if, once we have constructed them, there wouldn't be anything we could do with them. But of course there is. One thing we do with semantic representations we construct from linguistic input is to employ them as premises, usually in conjunction with representations we already have, and that often come from other sources (e.g. from what we have seen with our own eyes). In this way, the new representation may yield additional information . . .."
    (Foreword to Computational Semantics with Functional Programming by Jan van Eijck and Christina Unger. Cambridge University Press, 2010)