Word sense disambiguation with multilingual features

  • Authors:
  • Carmen Banea;Rada Mihalcea

  • Affiliations:
  • University of North Texas;University of North Texas

  • Venue:
  • IWCS '11 Proceedings of the Ninth International Conference on Computational Semantics
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper explores the role played by a multilingual feature representation for the task of word sense disambiguation. We translate the context of an ambiguous word in multiple languages, and show through experiments on standard datasets that by using a multilingual vector space we can obtain error rate reductions of up to 25%, as compared to a monolingual classifier.