Domain Independent Learning of Ontology Mappings

This paper proposes a domain independent method for handling interoperability problems by learning a mapping between ontologies. The learning method is based on ex-changing instances of concepts that are defined in the on-tologies. The method starts with identifying pairs of in-stances of concepts denoting the same entity in the world using information retrieval techniques, followed by propos-ing and evaluating mappings between the ontologies using the pairs of instances. For each step of this method, the like-lihood that a decision is correct is taken into account. Im-portant benefits of the method are that (a) no domain knowl-edge is required, and (b) the structures of ontologies be-tween which a mapping must be established, play no role. 1