A Profile Propagation and Information Retrieval Based Ontology Mapping Approach

Abstract

Ontology has been used widely to help finding relevant information among distributed and heterogeneous sources. Given that no universal ontology exists for the WWW, ontology mapping attracts many researchers' interest in various areas. In this paper we propose a new generic ontology mapping approach based on profile propagation and information retrieval techniques. The features used to establish the profile of a concept include all lexical information (i.e., its name, label, comments, property restriction, etc.). Profile propagation then is used to integrate structural information by adding the profiles of its ancestors, descendants and siblings to the profile of a concept, with different weights. Afterwards profiles are compared in a vector space model and the most relevant one will be returned as mapping results. A search engine is integrated as profile mapper when the size of ontologies is large. Experimental results show that the proposed approach obtained a good performance in OAEI campaign 2006.