Identifier les pronoms anaphoriques et trouver leurs antécédents : l'intérêt de la classification bayésienne

Davy Weissenbacher (1) , Adeline Nazarenko (1)

1 LIPN - Laboratoire d'Informatique de Paris-Nord

In NLP, a traditional distinction opposes linguistically-based systems and knowledge-poor ones, which mainly rely on surface clues. Each approach has its drawbacks and its advantages. In this paper, we propose a new approach based on Bayes Networks that allows to combine both types of information. As a case study, we focus on the anaphora resolution which is known as a difficult NLP problem. We show that our bayesain system performs better than a state-of-the-art one for this task.

Domaines: Traitement du texte et du document

Davy Weissenbacher, Adeline Nazarenko. Identifier les pronoms anaphoriques et trouver leurs antécédents : l'intérêt de la classification bayésienne. Traitement Automatique des Langues Naturelles (TALN'07), Jun 2007, France. pp.145-155. ⟨hal-00162093⟩