Multiple-sources informations fusion - a pratical inconsistency-tolerant approach

Didier Dubois (1, 2) , Hélène Fargier (1, 2) , Henri Prade (1, 2)

1 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage
2 CNRS - Centre National de la Recherche Scientifique

A multiple-sources information system should be able to provide the user with the most possibly accurate answers about the values of the attributes describing the objects of interest. In this paper, the attribute values, which are supposed to range on linearly ordered domains, may be imprecisely given by the sources, which more over may not be fully reliable. After a brief background on the general principles under lying the fusion of imprecise information in the framework of possibility theory, the case of single-valued attributes is first considered, before dealing with the multiple-valued case. The handling of known relations linking attributes, in a fusion process, is also briefly discussed.

Mots clés: fusion; possibility theory; inconsistency; maximal consistent subsets

Domaines: Intelligence artificielle [cs.AI]

Didier Dubois, Hélène Fargier, Henri Prade. Multiple-sources informations fusion - a pratical inconsistency-tolerant approach. 8th International Conference on Information Processing and Management of Uncertainty in Knowledge-based Systems (IPMU 2000), Jul 2000, Madrid, Spain. ⟨hal-03405306⟩