An Improved Ontology Mapping System Based on GRA

摘要

Ontology mapping created by end-users contains class hierarchy structure knowledge of the domain, and using that knowledge could produce new domainal ontology mapping. However, current systems have the following drawbacks: insufficient consideration of class hierarchy complexity, incomplete deletion inconsistent knowledge and inefficiency. Aiming at solving those problems, this paper proposes a new method of dealing differently with "similar to" and "equal to" relations in grass-roots ontology alignment. Besides, this paper re-designed inferences on class hierarchy knowledge base. The experiment result shows that revised system is more efficient than the original one, can produce results with higher precision while maintaining high recall, and can support complex ontology mapping.