Boosting applied to word sense disambiguation

Gerard Escudero, Lluis Marques, German Rigau

Research output: Chapter in Book/Report/Conference proceedingConference contribution

59 Citations (Scopus)

Abstract

In this paper Schapire and Singer’s AdaBoost.MH boosting algorithm is applied to the Word Sense Disambiguation (WSD)pro blem. Initial experiments on a set of 15 selected polysemous words show that the boosting approach surpasses Naive Bayes and Exemplar–based approaches, which represent state–of–the–art accuracy on supervised WSD. In order to make boosting practical for a real learning domain of thousands of words, several ways of accelerating the algorithm by reducing the feature space are studied. The best variant, which we call LazyBoosting, is tested on the largest sense–tagged corpus available containing 192,800 examples of the 191 most frequent and ambiguous English words. Again, boosting compares favourably to the other benchmark algorithms.

Original languageEnglish
Title of host publicationMachine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings
PublisherSpringer Verlag
Pages129-141
Number of pages13
Volume1810
ISBN (Print)9783540451648
Publication statusPublished - 2000
Externally publishedYes
Event11th European Conference on Machine Learning, ECML 2000 - Barcelona, Catalonia, Spain
Duration: 31 May 20002 Jun 2000

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1810
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other11th European Conference on Machine Learning, ECML 2000
CountrySpain
CityBarcelona, Catalonia
Period31/5/002/6/00

Fingerprint

Word Sense Disambiguation
Boosting
Adaptive boosting
Naive Bayes
AdaBoost
Ambiguous
Feature Space
Benchmark
Experiments
Experiment

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Escudero, G., Marques, L., & Rigau, G. (2000). Boosting applied to word sense disambiguation. In Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings (Vol. 1810, pp. 129-141). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1810). Springer Verlag.

Boosting applied to word sense disambiguation. / Escudero, Gerard; Marques, Lluis; Rigau, German.

Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings. Vol. 1810 Springer Verlag, 2000. p. 129-141 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1810).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Escudero, G, Marques, L & Rigau, G 2000, Boosting applied to word sense disambiguation. in Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings. vol. 1810, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1810, Springer Verlag, pp. 129-141, 11th European Conference on Machine Learning, ECML 2000, Barcelona, Catalonia, Spain, 31/5/00.
Escudero G, Marques L, Rigau G. Boosting applied to word sense disambiguation. In Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings. Vol. 1810. Springer Verlag. 2000. p. 129-141. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Escudero, Gerard ; Marques, Lluis ; Rigau, German. / Boosting applied to word sense disambiguation. Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings. Vol. 1810 Springer Verlag, 2000. pp. 129-141 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b3975e9819f846fc8fe12b31e3139ae4,
title = "Boosting applied to word sense disambiguation",
abstract = "In this paper Schapire and Singer’s AdaBoost.MH boosting algorithm is applied to the Word Sense Disambiguation (WSD)pro blem. Initial experiments on a set of 15 selected polysemous words show that the boosting approach surpasses Naive Bayes and Exemplar–based approaches, which represent state–of–the–art accuracy on supervised WSD. In order to make boosting practical for a real learning domain of thousands of words, several ways of accelerating the algorithm by reducing the feature space are studied. The best variant, which we call LazyBoosting, is tested on the largest sense–tagged corpus available containing 192,800 examples of the 191 most frequent and ambiguous English words. Again, boosting compares favourably to the other benchmark algorithms.",
author = "Gerard Escudero and Lluis Marques and German Rigau",
year = "2000",
language = "English",
isbn = "9783540451648",
volume = "1810",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "129--141",
booktitle = "Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings",

}

TY - GEN

T1 - Boosting applied to word sense disambiguation

AU - Escudero, Gerard

AU - Marques, Lluis

AU - Rigau, German

PY - 2000

Y1 - 2000

N2 - In this paper Schapire and Singer’s AdaBoost.MH boosting algorithm is applied to the Word Sense Disambiguation (WSD)pro blem. Initial experiments on a set of 15 selected polysemous words show that the boosting approach surpasses Naive Bayes and Exemplar–based approaches, which represent state–of–the–art accuracy on supervised WSD. In order to make boosting practical for a real learning domain of thousands of words, several ways of accelerating the algorithm by reducing the feature space are studied. The best variant, which we call LazyBoosting, is tested on the largest sense–tagged corpus available containing 192,800 examples of the 191 most frequent and ambiguous English words. Again, boosting compares favourably to the other benchmark algorithms.

AB - In this paper Schapire and Singer’s AdaBoost.MH boosting algorithm is applied to the Word Sense Disambiguation (WSD)pro blem. Initial experiments on a set of 15 selected polysemous words show that the boosting approach surpasses Naive Bayes and Exemplar–based approaches, which represent state–of–the–art accuracy on supervised WSD. In order to make boosting practical for a real learning domain of thousands of words, several ways of accelerating the algorithm by reducing the feature space are studied. The best variant, which we call LazyBoosting, is tested on the largest sense–tagged corpus available containing 192,800 examples of the 191 most frequent and ambiguous English words. Again, boosting compares favourably to the other benchmark algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84974712696&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84974712696&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783540451648

VL - 1810

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 129

EP - 141

BT - Machine Learning: ECML 2000 - 11th European Conference on Machine Learning, Proceedings

PB - Springer Verlag

ER -