Research On Concept Similarity Between OWL Ontology.

Ontology similarity measure has for a long time been a subject of intense scholarship in the fields of ontology mapping and integration. Concept similarity is the basis of ontology similarity. In the Semantic Web background, research areas such as ontology mapping, information integration and semantic search all need to recognize and estimate the semantic between different ontologies in order to share knowledge. However, currently, most exist technologies have some shortage and are not sufficient for applications because these methods are depending on syntactic but not implied semantic. How to mine hint semantic of ontology, dedicate to calculate similarity through knowledge expression and implement the algorithm base on semantic between concepts of different ontology are the important issues in the current ontology research area.Under the background stated above, the analysis of exist approaches of concept similarity are presented in this paper, proposing a special method using semantic expression of description logic. Basing on the approaches proposed by T.-L. BACH algorithm, this paper has stated a improved algorithm of concept similarity, to measure the semantic similarity between concept of different ontologies expressed by OWL DL. A prototype tool for it is designed and implemented, which provides the function of OWL ontology browsing and similarity calculation. Rank and Difference were introduced to evaluate the similarity result. Using these guidelines to compare OCSim Algorithm with T-L BACH Algorithm, the experiment result show that OCSim not only keep character of T-L BACH, but also better in distinguishing the similarity of concepts in the same level of semantic net, and Difference guideline result is more excellent. Case studies show that OCSim Algorithm have abundant expressiveness about concept similarity, more reflection of ontology designer and semantic character of ontology concepts. So OCSim Algorithm is effective and make sense for research area of ontology mapping.