Abstract

This paper describes an experimental comparison of seven different learning algorithms on the problem of learning to disambiguate the meaning of a word from context. The algorithms tested include statistical, neural-network, decision-tree, rule-based, and case-based classification techniques. The specific problem tested involves disambiguating six senses of the word "line" using the words in the current and proceeding sentence as context. The statistical and neural-network methods perform the best on this particular problem and we discuss a potential reason for this ob- served difference. We also discuss the role of bias in machine ]earning and its importance in explaining performance differences observed on specific problems.

Citations

4071	C4.5: Programs for machine learning - Quinlan - 1993
3341	Pattern Classification and Scene Analysis - Duda, Hart - 1973
2529	C.J.E.: Uci repository of machine learning databases, http://www.ics.uci.edu/ ∼ mlearn/MLRepository.html - Blake, Keogh
2304	Learning Internal Representations by Error Propagation - Rumelhart, Hinton, et al. - 1986
1654	B.: Building a large annotated corpus of english: The Penn treebank - Marcus, Marcinkiewicz, et al. - 1993
1302	WordNet: an online lexical database - Miller, Beckwith, et al. - 1990
897	Instance-Based Learning Algorithms - Aha, Kibler, et al. - 1991
784	Learning logical definitions from relations - Quinlan - 1989
687	Nearest neighbor pattern classification - Cover, Hart - 1967
682	The CN2 induction algorithm - Clark, Niblett - 1989
649	A stochastic parts program and noun phrase parser for unrestricted text - Church - 1988
637	A theory and methodology of inductive learning - Michalski - 1983
594	A vector space model for automatic indexing - Salton, Wong, et al. - 1975
496	Statistical Language Learning - Charniak - 1993
488	an algorithm for finding best matches in logarithmic expected time - Friedman, Bentley, et al. - 1977
385	Very simple classification rules perform well on most commonly used datasets - Holte - 1993
347	Learning decision lists - Rivest - 1987
265	Structural ambiguity and lexical relations - Hindle, Rooth - 1993
247	Pattern Classi cation and Scene Analysis - Duda, Hart - 1973
241	Principles of neurodynamics - Rosenblatt - 1962
222	Computer systems that learn: classification and prediction methods from statistics, neural nets, machine learning, and expert systems - Weiss, Kulikowski - 1991
200	A method for disambiguating word senses in a large corpus. Computers and the Humanities - Gale, Church, et al. - 1992
190	Boolean feature discovery in empirical learning - Pagallo, Haussler - 1990
174	Refinement of approximate domain theories by knowledge-based neural networks - Towell, Shavlik, et al. - 1990
167	The Need for Biases in Learning Generalizations - Mitchell - 1990
166	Word sense disambiguation using statistical methods - Brown, Pietra, et al. - 1991
140	A Conservation Law for Generalization Performance - Schaffer - 1994
126	Decision lists for lexical ambiguity resolution: application to accent restoration in Spanish and French.” 32nd Annual Meeting of the Association for Computational Linguistics - Yarowsky - 1994
122	An empirical comparison of pattern recognition, neural nets, and machine learning classi®cation methods - Weiss, Kapouleas - 1989
120	Automatic grammar induction and parsing free text: A transformationbased approach - Brill - 1993
104	Elements of Machine Learning - Langley - 1996
95	Symbolic and neural learning algorithms: An experimental comparison - Shavlik, Mooney, et al. - 1991
91	Estimating upper and lower bounds on the performance of word-sense disambiguation programs - Gale, Church, et al. - 1992
89	Generalizing from case studies: a case study - Aha - 1992
87	Evaluating automated and manual acquisition of anaphora resolution strategies - Aone, Bennett - 1995
82	An experimental comparison of symbolic and connectionist learning algorithms - Mooney, Shavlik, et al. - 1989
80	An Empirical Comparison of ID3 and Back-Propagation - Fisher, McKusick - 1986
79	Learning logical de nitions from relations - Quinlan - 1990
68	Induction of firstorder decision lists: results on learning the past tense of English verbs - Mooney, Califf - 1995
65	Corpus-based statistical sense resolution - Leacock, Towell, et al. - 1993
57	Selecting a classification method by cross-validation - Schaffer - 1993
55	On the Connection Between In-sample Testing and Generalization Error - Wolpert - 1992
53	ID2-of-3: constructive induction of M-of-N concepts for discriminators in decision trees - Murphy, Pazzani - 1991
45	A Comparative Study of ID3 and Backpropagation for English Text-to-speech - Dietterich, Hild, et al. - 1990
39	Answering the connectionist challenge: a symbolic model of learning the past tenses of English verbs - Ling, Marinov - 1993
38	For every generalization action, is there really an equal and opposite reaction? Analysis of the conservation law for generalization performance - RB, Gordon, et al. - 1995
35	Symbolic revision of theories with M-of-N rules - Baffes, Mooney - 1993
33	Learning the past tense of English verbs: The symbolic Pattern Associator vs. Connectionist Models - Ling - 1994
31	Evaluation and selection of biases in machine learning - Gordon, desJardins - 1995
25	An algorithm for �nding best matches in logarithmic expected time - Friedman, Bentley, et al. - 1977