Using Maximal Join for Information Fusion - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2009

Using Maximal Join for Information Fusion

Résumé

Information fusion is a very active research do-main. A lot of studies exist dealing with informa-tion fusion at a low level of semantics. Our claimis that information should be fused at a high levelof semantics and using a symbolic representation.Previously, we intuitively presented a framework for high-level symbolic fusion. Our approach relieson the use of the conceptual graphs model. Domainknowledge is a major point of the fusion process.The use of conceptual graphs for knowledge rep-resentation fusion eases the process of expressingdomain knowledge and fusion heuristics. In thispaper, we formalize our approach. In particular,we detail and formalize the introduction of domainknowledge inside the fusion process. We validateour approach within the context of a TV programrecommendation system.
Fichier non déposé

Dates et versions

hal-01294645 , version 1 (29-03-2016)

Identifiants

  • HAL Id : hal-01294645 , version 1

Citer

Claire Laudy, Jean-Gabriel Ganascia. Using Maximal Join for Information Fusion. FIRST IJCAI WORKSHOP ON GRAPH STRUCTURES FORKNOWLEDGE REPRESENTATION AND REASONING, Jul 2009, Pasadena, California, United States. ⟨hal-01294645⟩
139 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More