Possibilistic networks for information retrieval (IRGM 2007)

Mohand Boughanem (1) , Asma Brini (1) , Didier Dubois (2, 3)

1 IRIT-SIG - Systèmes d’Informations Généralisées
2 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage
3 CNRS - Centre National de la Recherche Scientifique

This paper proposes an information retrieval (IR) model based on possibilistic directed networks. The relevance of a document w.r.t a query is interpreted by two degrees: the necessity and the possibility. The necessity degree evaluates the extent to which a given document is relevant to a query, whereas the possibility degree evaluates the reasons of eliminating irrelevant documents. This new interpretation of relevance led us to revisit the term weighting scheme by explicitly distinguishing between informative and noninformative terms in a document. Experiments carried out on three standard TREC collections show the effectiveness of the model.

Mots clés: Information retrieval; Possibilistic networks; Bayesian networks; Relevance; Entropy

Domaines: Intelligence artificielle [cs.AI]

Mohand Boughanem, Asma Brini, Didier Dubois. Possibilistic networks for information retrieval (IRGM 2007). International ACM SIGIR Conference - Workshop Information retrieval and applications of graphical models (IRGM 2007), Jul 2007, Amsterdam, Netherlands. pp.67-89. ⟨hal-03445673⟩