HMMs, GRs, and n-grams as lexical substitution techniques: are they portable to other languages?

  • Authors:
  • Judita Preiss;Andrew Coonce;Brittany Baker

  • Affiliations:
  • The Ohio State University;The Ohio State University;The Ohio State University

  • Venue:
  • MCTLLL '09 Proceedings of the Workshop on Natural Language Processing Methods and Corpora in Translation, Lexicography, and Language Learning
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We introduce a number of novel techniques to lexical substitution, including an application of the Forward-Backward algorithm, a grammatical relation based similarity measure, and a modified form of n-gram matching. We test these techniques on the Semeval-2007 lexical substitution data [McCarthy and Navigli, 2007], to demonstrate their competitive performance. We create a similar (small scale) dataset for Czech, and our evaluation demonstrates language independence of the techniques.