A Method Of Top-k Ontology Matching

In recent years, ontology has become a hot topic in the fields of Artificial Intelligence, Knowledge Engineer, Semantic Web, Data Integration, Information Retrieval, and so on. The main purpose of applying ontology is knowledge sharing and reusing. Because there is not a common criterion for building ontology, a large amount of heterogeneous ontologies are constructed in different domains or even in the same domain. Ontology matching can solve the problem of ontological heterogeneity.In this dissertation, a method of Top-k ontology matching is proposed. The method is based on the computing of similarities between the elements. First, it calculates the similarity at element level, structure level and instance level according to the character of the heterogeneous ontologies. Then the integrated similarity of element pair is obtained by using combined strategy, in the similarity integration process. On the basis of taking into account the inherent uncertainty in the matching process, the dissertation select best k mapping in every iteration process rather than using a fixed threshold. Top-k ontology matching approach can deal with either 1: 1, 1: n, n: 1, or general m: n mapping. Another difference of the ontology matching approach form other approaches is that it can match the elements automatically without the intervention by users. At last, the realization and the validation show that the method can achieve efficiently matching between two ontologies.