earticle

영어

Ontology alignment finds matching elements from different ontologies. A number of ontology alignment techniques have been proposed. But few of them are adaptive to fuzzy ontologies alignment. A key problem is that formulas for computing similarity between elements from precise ontologies can’t be directly applied to those elements from fuzzy ontologies. In this paper, we exploit WordNet to compute similarity between two words. And Inspired by formula for computing similarity between two formal concepts from concept lattice, we propose an effective similarity formula for computing similarity between classes from different fuzzy ontologies. Also, we present an algorithm which is global optimal for aligning fuzzy ontologies. A prototype is implemented and the experiment shows that our approach is worthy.