Mapping validation by probabilistic reasoning

In the semantic web environment, where several independent ontologies are used in order to describe knowledge and data, ontologies have to be aligned by defining mappings among the elements of one ontology and the elements of another ontology. Very often, mappings are not derived from the semantics of the ontologies that are compared. Rather, mappings are computed by evaluating the similarity of the ontologies terminology and/or of their syntactic structure. In this paper, we propose a new mapping validation approach. The approach is based on the notion of probabilistic mappings and on the use of probabilistic reasoning techniques to enforce a semantic interpretation of similarity mappings as probabilistic and hypothetical relations among ontology elements.