Comparisons of Relatedness Measures through a Word Sense Disambiguation Task

Abstract : Michael Zock's work has focussed these last years on finding the appro- priate and most adequate word when writing or speaking. The semantic relatedness between words can play an important role in this context. We present here a novel approach to analyse the semantic relatedness between words based on the relevance of semantic relatedness measures on the global level of a word sense disambigua- tion task. We introduce two metrics : the best atteignable score and the correlation between global score and the F1 measure. We use them to analyse several classical local semantic similarity measures as well as measures built by our team.
Document type :
Book sections
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01003000
Contributor : Didier Schwab <>
Submitted on : Sunday, June 8, 2014 - 5:05:38 PM
Last modification on : Thursday, February 21, 2019 - 4:30:02 PM

Identifiers

  • HAL Id : hal-01003000, version 1

Collections

Citation

Didier Schwab, Andon Tchechmedjiev, Jérôme Goulian, Gilles Sérasset. Comparisons of Relatedness Measures through a Word Sense Disambiguation Task. Advances in Language Production, Cognition and the Lexicon,, Springer, pp.23, 2014. ⟨hal-01003000⟩

Share

Metrics

Record views

165