26 Junio 2007
University of South Australia
Dennis Hooijmaijers

The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries [w3c]. One of the main issues of the Semantic Web is mapping ontologies that may be subjective and created with bias toward a specific goal. In this work we propose integrating description logics with subjective logics to predict the likelihood that two ontologies use a concept in exactly the same way. By including subjective logics within an ontology the ability to capture belief, disbelief and uncertainty allows for a probabilistic estimate, that captures the level of similarity on terminology elucidation. To discover an a priori, a probability estimate used in the absence of evidence, currently we are investigating social networks to capture cliques whose common interest create an informal agreement on the elucidation of a term.