Languages & Linguistics
Computational Linguistics
Computational Linguistics is a field that combines linguistics and computer science to develop algorithms and models for understanding and processing natural language. It involves tasks such as speech recognition, language translation, and natural language processing. Computational linguists use techniques from machine learning, statistics, and linguistics to analyze and generate human language.
Written by Perlego with AI-assistance
Related key terms
1 of 5
10 Key excerpts on "Computational Linguistics"
eBook - PDFIntroducing Linguistics
Theoretical and Applied Approaches
- Joyce Bruhn de Garavito, John W. Schwieter(Authors)
- 2021(Publication Date)
- Cambridge University Press(Publisher)
Ambiguity causes a rapid increase in the number of possible inter- pretations of a natural language sentence. One method for dealing with this ambiguity was examined. Machine learning was mentioned as an alternative approach. Applications of Computational Linguistics abound, a few of which have been men- tioned in this overview. The rate at which new applications are developed and the increasing sophistication of the linguistic techniques that are incorporated into them is impressive. Computational Linguistics is part of the technological revolution that is changing how we interact with our world. 592 Computational Linguistics FURTHER READING For those of you who are interested in exploring Computational Linguistics more deeply, a number of books have been written. Here are a selected few: Bird, S., Klein, E., & Loper, E. (2009). Natural language processing with Python. Sebastopol, CA: O’Reilly Media. Jurafsky, D., & Martin, J. (2008). Speech and language processing (2 nd Edition). Upper Saddle River, NJ: Pearson. Manning, C., & Schütze, H. (1999). Foundations of statistical natural language processing. Cambridge, MA: MIT Press. Manning, C., Raghavan, P., & Schütze, H. (2008). Introduction to information retrieval. Cambridge, UK: Cambridge University Press. Speech Recognition Breakthrough for the Spoken, Translated Word: https://www.youtube.com/watch?v=Nu-nlQqFCKg OVERVIEW In this chapter, you will develop an understanding of how English came to be a world language historically and what types of English there are spoken around the world. Our objectives are to: • describe and compare models of English as a world language; • understand how Englishes are differentiated from first, second, and foreign language varieties; • introduce a typology of Englishes; and • understand what linguistic, social, and sociolinguistic factors are at play.
eBook - PDFWho Climbs the Grammar-Tree
[leaves for David Reibel]
- Rosemarie Tracy(Author)
- 2011(Publication Date)
- De Gruyter(Publisher)
507 Franz Guenthner Computational Linguistics: A Personal View 1 1. If Computational Linguistics is the answer, what was the question? I would like to start with a few notes on the name of this (relatively new) discipline, with some examples the tasks this discipline has addressed in the recent past, which are not necessarily inherent in the name 'Computational Linguistics 1 . This paper has three parts: first, I would like to make a suggestion concerning a possible basic predicate of this discipline, and how this predicate could be specified so as to partitition the discipline into sub-areas, each with its own specific task, i.e. its specific predicate. Then, I will use this predicate to characterize three goals of Computational Linguistics that I consider to be of central importance. At the end of this note, I will make some remarks about the role of the programming language PROLOG in the computational processing of the predicate and in the concrete realization of the above goals. 2. Computational Linguistics: 'Computable Linguistics'? Computational Linguistics is a young discipline with many names 2 and to this day a discipline with almost as many different methodological and theoretical assumptions as there are practicioners. How are we to understand 'computational 1 ? Two interpretations - at least - are possible: one addresses the concrete utilisation of computers and computer technology, the others involves concepts like 'enumerable' or 'programmable 1 . Both interpretations are, of course, related to each other in very direct ways. But as we shall see, it is the computability aspects that make the discipline interesting and it is the question of what exactly we want to compute in linguistics that this note deals with.
eBook - ePub- Dov M. Gabbay, Paul Thagard, John Woods(Authors)
- 2012(Publication Date)
- North Holland(Publisher)
inter alia, Computational Linguistics should engage in (p. 31):1. “basic developmental research in computer methods for handling language, as tools to help the linguistic scientist discover and state his generalizations, and as tools to help check proposed generalizations against data;” and2. “developmental research in methods to allow linguistic scientists to use computers to state in detail the complex kinds of theories (for example, grammars and theories of meaning) they produce, so that the theories can be checked in detail.“In very broad strokes, these describe an undercurrent that remained very influential in Computational Linguistics research until the mid-1990s, and that placed the field of Computational Linguistics necessarily in very close proximity to the contemporary research programmes of theoretical linguistics.2 Narratives of Progress
Then something happened. The post-1995 ACL community have woven a number of somewhat fictional narratives about what that something was, but the result has very clearly been a redefinition of Computational Linguistics that is much more answerable to its potential applications (including, very prominently, machine translation systems) and much less beholden to other branches of linguistics for its direction.2.1 The Advent of Statistical Methods
Perhaps the growing rift with linguistics can be felt most prominently today through the preference for statistical techniques in Computational Linguistics, a trend that generative linguists have been very slow to embrace. In CL, “statistical” is used to refer to both statistical sampling methods as well as probabilistic models. It will be used here to refer to the larger-dimensional statistical methods that are pervasive in engineering and statistical pattern recognition, as opposed to the statistics that are widely used in the life sciences and elsewhere for descriptive hypothesis testing. Indeed, quantitative approaches to experimental design and significance testing are not taught in Computational Linguistics curricula, have only begun to appear in CL publications within the last ten years, and remain poorly understood by most CL researchers.
eBook - PDF- Robert Lado, Norman A. McQuown, Sol Saporta, Robert Lado, Norman A. McQuown, Sol Saporta(Authors)
- 2019(Publication Date)
- De Gruyter Mouton(Publisher)
Computational Linguistics LEONARDO MANRIQUE CASTAÑEDA A more accurate, if more cumbersome, expression for 'Computational Linguistics' would be: 'the handling of linguistic material with the aid of electronic computers'. In reference to both pure and applied linguistcs, electronic computers have been used all over the world. Since it is necessary to formulate the programs with great attention to detail, work with computers has led to the formalization of many of the assumptions underlying linguistic research. Sometimes problems were clarified through new formulations of linguistic theories (e.g. the transformational theory), in other instances totally new theories and formulations had to be devised in order to adjust the data to the requirements of the input, processing, and output of computers. By way of illustration, we will mention a few of the applications of computers in linguistic research. First of all, we must discuss the attempts made in the field of machine translation. So far, the proposed end has not been attained, but the project has stimulated a great deal of research. There are now many descriptive studies available, in which the analysis of a language (the source) is made in accordance with a supposed algorithm of translation into another language (the target); or, when it is assumed that an intermediate stage will save time and effort, the analysis is made with this intermediate target in mind. The analysis of languages for machine translation has imposed new criteria for the recognition of roots, stems, and affixes, since a traditional definition has proved, in most instances, not suitable for computer processing. Computers have led to the formulation of new kinds of dictionaries, formulas of syntactic constructions, and the like, but we cannot examine all of them here. If machine translations are not yet possible, research has led, in several instances, and with varying degrees of success, to 'machine aided translations'.
eBook - PDF- David Golumbia(Author)
- 2009(Publication Date)
- Harvard University Press(Publisher)
Contemporary CL and NLP projects, pro-ceed along several different lines, some inspired by the work originally done by Weaver and other MT advocates, and other work inspired by the use of computers by linguists. Contemporary CL and NLP includes at least the following areas of inquiry: text-to-speech synthesis (TTS): the generation Computationalist Linguistics p 93 of (synthesized) speech from written text; voice recognition: the use of human language for input into computers, often as a substitute for written/-keyboarded text; part-of-speech tagging: the assignment of grammatical categories and other feature information in human language texts; corpus linguistics: the creation and use of large, computer-based collections of texts, both written and spoken; natural language generation (NLG) or speech production: the use of computers to spontaneously emit human language; conversational agents: the computer facility to interact with hu-man interlocutors, but without the ability to spontaneously generate speech on new topics; CL of human languages: the search for computa-tional elements in human languages; statistical NLP: the analysis of hu-man language (often large corpora) using statistical methods; information extraction: the attempt to locate the informational content of linguistic ex-pressions and to synthesize or collect them from a variety of texts. The vis-ible success in some of these programs (especially ones centered around speech synthesis and statistical analysis, neither of which has much to do with what is usually understood as comprehension; Manning and Schütze [1999] provides a thorough survey) leads to a popular misconception that other programs are on the verge of success.
eBook - PDFCognitive Linguistics
Current Applications and Future Perspectives
- Gitte Kristiansen, Michel Achard, René Dirven, Francisco J. Ruiz de Mendoza Ibáñez, Gitte Kristiansen, Michel Achard, René Dirven, Francisco J. Ruiz de Mendoza Ibáñez(Authors)
- 2008(Publication Date)
- De Gruyter Mouton(Publisher)
It is indeed the case that such applications are cur-rently in fast expansion in many fields, as the chapters just surveyed in this introduction amply show. The CL usage-based approach furthermore naturally calls for a non-restrictive interpretation of the term ‘appli-cation’: one of the main assumptions of ACL is that linguistic research must be firmly grounded in usage-based, descriptive data, which may, more often than not, imply an interdisciplinary approach to the object under scrutiny. In language pedagogy, for instance, linguistics meets with psycholinguistics and educational psychology to study the combined pro-cesses of natural and conducted acquisition of second and/or foreign lan-guages; hence the label ‘pedagogical linguistics’ is out of place. In cultural linguistics, the deeper relation between language and culture is explored, which covers and overlaps with various scientific disciplines or fields such as (cognitive) anthropology, cross-cultural semantics, intercultural prag-matics, cross-cultural communication, social psychology and still other disciplines. In Computational Linguistics, amongst many other possibil-ities, theoretical models such as cognitive grammar, conceptual metaphor theory, blending theory, etc. are caught in self-monitoring artificial intel-ligence programs to test the viability of the constructs used. As to the scope of ACL the following may serve as an attempt to spell out some examples of ACL’s orientations, at best signposting some major areas on the map, but certainly not laying out a complete map. One major application of CL is the area of descriptive linguistics cover-ing either the description of parts or the whole of the grammar of given languages, or else the description of language as used in given domains of human experience.
- Richard A. Geiger, Brygida Rudzka-Ostyn, Richard A. Geiger, Brygida Rudzka-Ostyn(Authors)
- 2011(Publication Date)
- De Gruyter Mouton(Publisher)
Before I begin the presentation of process linguistics via a short intro-duction to cognitive science, a brief note about its exact nature is in or-der. I want to stress that PL is NOT a linguistic theory as it stands, but rather a framework of principles and notions within which a (growing) number of existing (especially computational) approaches to natural language can be accommodated. Several other researchers have ex-pressed the same ideas in different ways, especially Winograd (1977, 1983), when he introduced the computational paradigm for the study of language. If bringing these principles and notions together can be a 142 Geert Adriaens stimulus for other researchers to work along the same lines, then the main aim of this presentation will have been reached. 2. Cognitive science 2.1. A new paradigm Cognitive science 1 is a contemporary scientific paradigm that is attempting to bring together a number of existing fields (artificial intelligence, psychology, neuroscience, philosophy, linguistics and anthropology) in a concerted effort to study the complex domain of cognition/intelligence in its broadest sense (including, for example, problems of knowledge representation, language processing, learning, reasoning and problem solving). To reach this goal it uses the research tools recently developed in its participating sciences. Its most important tools come from artificial intelligence (computer simulation of theories, using a wide variety of computer languages and formalisms) and cognitive psychology (rigorous experimentation and disciplined introspection (see Simon & Ericsson 1984)). So a full-scale cognitive-scientific study of natural language processing should fulfill the requirements of computational realizability and psychological reality: a model/theory should be accompanied by a computer program simulating its workings, and embedded within corroborating research findings from psycholinguistic experiments.
eBook - PDF- Mark Aronoff, Janie Rees-Miller(Authors)
- 2008(Publication Date)
- Wiley-Blackwell(Publisher)
Nonetheless, however confusing parts of this field may at times appear to be, this research is clearly in the direction of a more detailed and accurate understanding of language disorders and the neural mechanisms that support normal language knowledge and use. NOTE This work was supported by a grant from the National Institute for Neurological Dis- ease and Stroke (DC00942). 608 Sproat, Samuelsson, Chu-Carroll, and Carpenter 25 Computational Linguistics RICHARD SPROAT, CHRISTER SAMUELSSON, JENNIFER CHU-CARROLL, and BOB CARPENTER The field of Computational Linguistics is as diverse as linguistics itself, so giv- ing a thorough overview of the entire field in the short space available for this chapter is essentially impossible. We have therefore chosen to focus on four relatively popular areas of inquiry: • syntactic parsing; • discourse analysis; • computational morphology and phonology; • corpus-based methods. The order of presentation is motivated by historical considerations. Parsing and discourse analysis have had the longest continuous history of investigation, and are therefore presented first. Computational morphology and phonology only really began to grow as a separate discipline in the mid-1980s. Corpus- based approaches were, in fact, investigated as early as the 1960s (e.g., by Zellig Harris (1970)), but the field fell into disrepute until the late 1980s, since which time there has been a renaissance of work in this area. 1 Parsing Parsing is the act of determining the “syntactic structure” of a sentence. Although syntactic theories differ on their notions of structure, the goal of such structure is typically to represent “who did what to whom” in the sentence. Any natural language processing system that needs to produce an interpretation from the utterance that is deeper than a bag of keywords thus involves some form Computational Linguistics 609 of parsing (see section 1.5 for examples of existing practical applications of parsing).
eBook - PDF- Robert Freidin(Author)
- 2020(Publication Date)
- Cambridge University Press(Publisher)
This investigation also demonstrates how the interpretation of words – the 3 nouns in the title – is inexorably tied to our knowledge of the world. It is a virtual certainty that someone who has not read this chapter (or taken a linguistics course or read about modern linguistics) will not interpret language and linguistics as the chapter does. The computational system contains the mechanisms that derive the struc- tured expressions of languages from the contents of the lexicon. One mechan- ism Merge applies to the elements of the lexicon, creating syntactic units which specify the hierarchical syntactic structure of linguistic expressions (titles of books and courses, and of course sentences). In addition, there is a mechanism Linearize that creates a linear order of the words for these hier- archical structures. The distinction between the two mechanisms now under- pins the demonstration in the previous chapter of how hierarchical structure, but not linear order, determines interpretation. This is further supported in this chapter by the comparison of word order in English and Japanese, where expressions containing words with corresponding interpretations have differ- ent linear orders but the same hierarchical structure, yielding the same inter- pretation. And finally, the computational system contains a mechanism for labeling syntactic units (Label), which distinguishes, for example, between a subject the striped fish and a predicate swam lazily upstream in the sentence the striped fish swam lazily upstream by labeling the subject as N and the predicate as V. The definition of a language as a mental lexicon plus a computational system in the mind of the speaker provides a solid basis for defining language. From this perspective, language is what languages share in common: general properties of the computational system and the lexicon, properties that appear Coda 51
- Olga Akhmanova(Author)
- 2019(Publication Date)
- De Gruyter Mouton(Publisher)
It is time now to make it quite plain that the sketchy and sum-mary account of the now well known approaches to Computational Linguistics was included in this book because of the very import -and role it plays in the optimisation of natural communication systems. Computational Linguistics has done very much, and will do even more, to help solve some of the more difficult lexico-graphical problems. As soon as we come to a general agreement on the basic taxonomies of information retrieval, we shall be able to rely to an ever increasing extent on the automata for our knowledge of literary data. But in the present context the most important aspect is, of course, a rigorous analysis of natural communicative systems. What is the relationship, in each par-ticular case, between compilation, systematization and simu-lation? Is logical systematization at all possible on the higher levels of language ? Is there any hope for rational semantics or shall it, if not forever at least in the foreseeable future, be con-fined to the very limited set of the obvious cases - cases like, for instance, terms of relationship or military ranks where the sum-total of semes is easily deducible without any improved electronic hardware ? The most firmly established achievement of Computational Linguistics is the use of statistics, the boosting of quantitative linguistics into the position of one of the most important subdiv-isions of our science. In the next chapter we shall cast a glance at some of the achievements, as well as problems which still re-main to be solved. What do we count and why ? As one of the out-standing Soviet mathematicians, Academician Kolmogorov said in one of his lectures on applied mathematics, before we begin to count we must know what it is we are going to count and for what purpose. One did have the uneasy feeling that these obvious prerequisites were not always present in the minds of some of
Index pages curate the most relevant extracts from our library of academic textbooks. They’ve been created using an in-house natural language model (NLM), each adding context and meaning to key research topics.
