PageRank-based Word Sense Induction within Web Search Results Clustering

José G. Moreno (1) , Gaël Dias (1)

1 Equipe Hultech - Laboratoire GREYC - UMR6072

Word Sense Induction is an open problem in Natural Lan- guage Processing. Many recent works have been addressing this problem with a wide spectrum of strategies based on content analysis. In this paper, we present a sense induction strategy exclusively based on link analysis over the Web. In particular, we explore the idea that the main di erent senses of a given word share similar linking properties and can be found by performing clustering with link-based similarity metrics. The evaluation results show that PageRank-based sense induction achieves interesting results when compared to state-of-the-art content-based algorithms in the context of Web Search Results Clustering.

Mots clés: Word Sense Indiction; Web Links; PageRank Clustering

Domaines: Traitement du texte et du document; Apprentissage [cs.LG]

José G. Moreno, Gaël Dias. PageRank-based Word Sense Induction within Web Search Results Clustering. Joint Conference on Digital Libraries (JCDL 2014), Sep 2014, Londres, United Kingdom. 2 p. ⟨hal-01070313⟩