Word sense disambiguation in evolutionary manner

Saad Adnan Abed, Sabrina Tiun, Nazlia Omar

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The task of assigning proper meaning to an ambiguous word in a particular context is termed word sense disambiguation (WSD). We propose a genetic algorithm, improved by local search techniques, to maximise the overall semantic similarity or relatedness of a given text. Local search is used because of the inefficiency of population-based algorithms (e.g. genetic algorithm) in exploiting the search space. Firstly, the proposed method assigns all potential senses for each word using a WordNet sense inventory. Then, the improved genetic algorithm is applied to determine a coherent set of senses that carries maximum similarity or relatedness score based on information content and gloss overlap methods, namely extended Lesk algorithm and Jiang and Conrath (jcn). The obtained results outperformed other unsupervised methods, which are related to the proposed method, when tested on the same benchmark dataset. It can be concluded that the proposed method is an effective solution for unsupervised WSD.

Original languageEnglish
Pages (from-to)1-16
Number of pages16
JournalConnection Science
DOIs
Publication statusAccepted/In press - 13 Feb 2016

Fingerprint

Genetic algorithms
Semantics
Local search (optimization)

Keywords

  • genetic algorithm
  • local search
  • semantic relatedness
  • semantic similarity
  • Word sense disambiguation
  • WordNet

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Software

Cite this

Word sense disambiguation in evolutionary manner. / Adnan Abed, Saad; Tiun, Sabrina; Omar, Nazlia.

In: Connection Science, 13.02.2016, p. 1-16.

Research output: Contribution to journalArticle

@article{5abc570eaf2c4361a2e24d5c89e822fd,
title = "Word sense disambiguation in evolutionary manner",
abstract = "The task of assigning proper meaning to an ambiguous word in a particular context is termed word sense disambiguation (WSD). We propose a genetic algorithm, improved by local search techniques, to maximise the overall semantic similarity or relatedness of a given text. Local search is used because of the inefficiency of population-based algorithms (e.g. genetic algorithm) in exploiting the search space. Firstly, the proposed method assigns all potential senses for each word using a WordNet sense inventory. Then, the improved genetic algorithm is applied to determine a coherent set of senses that carries maximum similarity or relatedness score based on information content and gloss overlap methods, namely extended Lesk algorithm and Jiang and Conrath (jcn). The obtained results outperformed other unsupervised methods, which are related to the proposed method, when tested on the same benchmark dataset. It can be concluded that the proposed method is an effective solution for unsupervised WSD.",
keywords = "genetic algorithm, local search, semantic relatedness, semantic similarity, Word sense disambiguation, WordNet",
author = "{Adnan Abed}, Saad and Sabrina Tiun and Nazlia Omar",
year = "2016",
month = "2",
day = "13",
doi = "10.1080/09540091.2016.1141874",
language = "English",
pages = "1--16",
journal = "Connection Science",
issn = "0954-0091",
publisher = "Taylor and Francis AS",

}

TY - JOUR

T1 - Word sense disambiguation in evolutionary manner

AU - Adnan Abed, Saad

AU - Tiun, Sabrina

AU - Omar, Nazlia

PY - 2016/2/13

Y1 - 2016/2/13

N2 - The task of assigning proper meaning to an ambiguous word in a particular context is termed word sense disambiguation (WSD). We propose a genetic algorithm, improved by local search techniques, to maximise the overall semantic similarity or relatedness of a given text. Local search is used because of the inefficiency of population-based algorithms (e.g. genetic algorithm) in exploiting the search space. Firstly, the proposed method assigns all potential senses for each word using a WordNet sense inventory. Then, the improved genetic algorithm is applied to determine a coherent set of senses that carries maximum similarity or relatedness score based on information content and gloss overlap methods, namely extended Lesk algorithm and Jiang and Conrath (jcn). The obtained results outperformed other unsupervised methods, which are related to the proposed method, when tested on the same benchmark dataset. It can be concluded that the proposed method is an effective solution for unsupervised WSD.

AB - The task of assigning proper meaning to an ambiguous word in a particular context is termed word sense disambiguation (WSD). We propose a genetic algorithm, improved by local search techniques, to maximise the overall semantic similarity or relatedness of a given text. Local search is used because of the inefficiency of population-based algorithms (e.g. genetic algorithm) in exploiting the search space. Firstly, the proposed method assigns all potential senses for each word using a WordNet sense inventory. Then, the improved genetic algorithm is applied to determine a coherent set of senses that carries maximum similarity or relatedness score based on information content and gloss overlap methods, namely extended Lesk algorithm and Jiang and Conrath (jcn). The obtained results outperformed other unsupervised methods, which are related to the proposed method, when tested on the same benchmark dataset. It can be concluded that the proposed method is an effective solution for unsupervised WSD.

KW - genetic algorithm

KW - local search

KW - semantic relatedness

KW - semantic similarity

KW - Word sense disambiguation

KW - WordNet

UR - http://www.scopus.com/inward/record.url?scp=84958554764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958554764&partnerID=8YFLogxK

U2 - 10.1080/09540091.2016.1141874

DO - 10.1080/09540091.2016.1141874

M3 - Article

SP - 1

EP - 16

JO - Connection Science

JF - Connection Science

SN - 0954-0091

ER -