Exploiting association rules and ontology for semantic document indexing

Fatiha Boubekeur-Amirouche (1) , Mohand Boughanem (2, 3) , Lynda Tamine (2)

1 UMMTO - Université Mouloud Mammeri [Tizi Ouzou]
2 IRIT-SIG - Systèmes d’Informations Généralisées
3 UT3 - Université Toulouse III - Paul Sabatier

This paper describes a novel approach for document indexing based on the discovery of contextual semantic relations between concepts. The concepts are first extracted from WordNet ontology. Then we propose to extend and to use the association rules technique in order to discover conditional relations between concepts. Finally, concepts and related contextual relations are organized into a conditional graph.

Mots clés: Conceptual indexing; Association rules; Information retrieval; WordNet

Domaines: Recherche d'information [cs.IR]

Fatiha Boubekeur-Amirouche, Mohand Boughanem, Lynda Tamine. Exploiting association rules and ontology for semantic document indexing. International Conference on Information Processing and Management of Uncertainty in Knowledge-based Systems (IPMU 2008), Jun 2008, Malaga, Spain. pp.464--472. ⟨hal-00476487⟩