Unsupervised word sense disambiguation rivaling supervised methods
Published 1 January 1995Open access
David Yarowsky
Citations2,408
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
An unsupervised learning algorithm for sense disambiguation that, when trained on unannotated English text, rivals the performance of supervised techniques that require time-consuming hand annotations.
Abstract
This paper presents an unsupervised learning algorithm for sense disambiguation that, when trained on unannotated English text, rivals the performance of supervised techniques that require time-consuming hand annotations. The algorithm is based on two powerful constraints---that words tend to have one sense per discourse and one sense per collocation---exploited in an iterative bootstrapping procedure. Tested accuracy exceeds 96%.
Keywords
Computer Science
Journal of the Royal Statistical Society Series B (Statistical Methodology)Maximum Likelihood from Incomplete Data Via the <i>EM</i> Algorithm
49,657 Citations1977A. P. Dempster, N. M. Laird +1 more
Digital Access to Scholarship at Harvard (DASH) (Harvard University)Maximum Likelihood from Incomplete Data via the EM Algorithm
4,523 Citations1977Dempster, Arthur P., Laird, Nan M. +1 more
International Journal of LexicographyIntroduction to WordNet: An On-line Lexical Database<sup>*</sup>
4,207 Citations1990George A. Miller, Richard Beckwith +3 more
Standard alphabetical procedures for organizing lexical information put together words that are spelled alike and scatter words with similar or related meanings haphazardly through the list.
Automatic sense disambiguation using machine readable dictionaries
1,756 Citations1986Michael Lesk
This procedure uses available dictionaries, so that it will process any text; and uses solely the immediate context to decide which sense of a word is intended (in written English).
Medical Entomology and ZoologyInference and disputed authorship : The Federalist
916 Citations1964Frederick Mosteller, David L. Wallace
Machine LearningLearning Decision Lists
761 Citations1987Ronald L. Rivest
This paper introduces a new representation for Boolean functions, called decision lists, and shows that they are efficiently learnable from examples, and strictly increases the set of functions known to be polynomially learnable, in the sense of Valiant (1984).
Word-sense disambiguation using statistical models of Roget's categories trained on large corpora
670 Citations1992David Yarowsky
A program that disambiguates English word senses in unrestricted text using statistical models of the major Roget's Thesaurus categories, enabling training on unrestricted monolingual text without human intervention.
Computers and the HumanitiesA method for disambiguating word senses in a large corpus
534 Citations1992William A. Gale, Kenneth Church +1 more
The proposed method was designed to disambiguate senses that are usually associated with different topics using a Bayesian argument that has been applied successfully in related tasks such as author identification and information retrieval.
One sense per collocation
384 Citations1993David Yarowsky
This paper shows that for certain definitions of collocation, a polysemous word exhibits essentially only one sense per collocation and utilizes this property in a disambiguation algorithm that achieves precision of 92% using combined models of very local context.
Word-sense disambiguation using statistical methods
369 Citations1991Peter F. Brown, Stephen A. Della Pietra +2 more
A statistical technique for assigning senses to words is described, which incorporated into the statistical machine translation system the error rate of the system decreased by thirteen percent.
Decision lists for lexical ambiguity resolution
362 Citations1994David Yarowsky
This paper presents a statistical decision procedure for lexical ambiguity resolution that exploits both local syntactic patterns and more distant collocational evidence, generating an efficient, effective, and highly perspicuous recipe for resolving a given ambiguity.
Machine LearningLearning decision lists
355 Citations1987Ronald L. Rivest
This paper introduces a new representation for Boolean functions, called decision lists, and shows that they are efficiently learnable from examples, and strictly increases the set of functions known to be polynomially learnable, in the sense of Valiant (1984).
ScholarlyCommons (University of Pennsylvania)A corpus-based approach to language learning
280 Citations1993Eric Brill
A learning algorithm is described that takes a small structurally annotated corpus of text and a larger unannotated corpus as input and automatically learns how to assign accurate structural descriptions to sentences not in the training corpus.
Word sense disambiguation using a second language monolingual corpus
268 Citations1994Ido Dagan, Alon Itai
Word-sense disambiguation using decomposable models
206 Citations1994Rebecca Bruce, Janyce Wiebe
This paper describes a method for formulating probabilistic models that use multiple contextual features for word-sense disambiguation, without requiring untested assumptions regarding the form of the model.
Word sense disambiguation with very large neural networks extracted from machine readable dictionaries
160 Citations1990Jean Véronis, Nancy Ide
This paper describes a means for automatically building very large neural networks (VLNNs) from definition texts in machine-readable dictionaries, and demonstrates the use of these networks for word sense disambiguation.
Corpus-based statistical sense resolution
155 Citations1993Claudia Leacock, Geoffrey G. Towell +1 more
Three corpus-based statistical sense resolution methods studied here attempt to infer the correct sense of a polysemous word by using knowledge about patterns of word cooccurrences, based on Bayesian decision theory, neural, networks, and content vectors.
International Conference on Acoustics, Speech, and Signal ProcessingA stochastic parts program and noun phrase parser for unrestricted text
127 Citations2003Kenneth Church
Subject-dependent co-occurrence and word sense disambiguation
88 Citations1991Joe Guthrie, Louise Guthrie +2 more
Using the subject classifications given in the machine-redable version of Longman's Dictionary of Contemporary English, subject-dependent co-occurrence links between words of the defining vocabulary are established to construct "neighborhoods" and the application of these neighborhoods to information retrieval is described.
arXiv (Cornell University)Decision Lists for Lexical Ambiguity Resolution: Application to Accent Restoration in Spanish and French
71 Citations1994David Yarowsky
IBM Journal of Research and DevelopmentAn experiment in computational discrimination of English word senses
69 Citations1988Ezra Black
Experimental results are presented which suggest that people can consistently determine in which of several given senses a word is being used in text, simply by examining the half dozen or so words just before and just after the word in focus.
Discrimination Decisions for 100,000-Dimensional Spaces
35 Citations1994William A. Gale, Kenneth Church +1 more
Annals of Operations ResearchDiscrimination decisions for 100,000-dimensional spaces
13 Citations1995William A. Gale, Kenneth Church +1 more
A method designed for the sense discrimination problems mentioned is introduced and areas for research based on observed shortcomings of the method are discussed, including the need for a robust version of this method.
National Conference on Artificial IntelligenceToward the essential nature of statistical knowledge in sense resolution
11 Citations1994Jill Fain Lehman
It is shown how viewing this residual representation of (word sense) pairs as a form of episodic memory can enable symbolic, knowledge-rich systems to take advantage of this source of regularity in performing sense resolution.