Uncertainty in Ontologies: Dempster-Shafer Theory for Data Fusion Applications

Amandine Bellenger (1) , Sylvain Gatepaille (2)

1 MIND - LITIS - Equipe Multi-agent, Interaction, Décision
2 EADS, Defence and Security

Nowadays ontologies present a growing interest in Data Fusion applications. As a matter of fact, the ontologies are seen as a semantic tool for describing and reasoning about sensor data, objects, relations and general domain theories. In addition, uncertainty is perhaps one of the most important characteristics of the data and information handled by Data Fusion. However, the fundamental nature of ontologies implies that ontologies describe only asserted and veracious facts of the world. Different probabilistic, fuzzy and evidential approaches already exist to fill this gap; this paper recaps the most popular tools. However none of the tools meets exactly our purposes. Therefore, we constructed a Dempster-Shafer ontology that can be imported into any specific domain ontology and that enables us to instantiate it in an uncertain manner. We also developed a Java application that enables reasoning about these uncertain ontological instances.

Mots clés: data fusion

Mots clés: ontologies; uncertainty; belief functions

Domaines: Modélisation et simulation; Web; Intelligence artificielle [cs.AI]

Amandine Bellenger, Sylvain Gatepaille. Uncertainty in Ontologies: Dempster-Shafer Theory for Data Fusion Applications. Workshop on Theory of Belief Functions, Apr 2010, Brest, France. pp.Bellenger. ⟨hal-00601667⟩