Improved Algorithm of Name Strategy and Structure Strategy in Ontology Matching
Jianer Chen · Journal of Chinese Computer Systems · 2010
This paper analyzes the main drawbacks of traditional name-based and structure-based strategies.To solve the problems of name-based strategies,a new solution based on WordNet is proposed.This solution maps core words of a concept to the WordNet to eliminate word ambiguity,using this approach will improve the accuracy of similarity calculation between words,and a C-C principle is adopted to solve the blindness problem which usually exists when combining words similarity values.To solve the problems of traditional structure-based strategies,a new algorithm of similarity propagation is proposed,which is based on the information content of concept that is obtained from the structure information of ontology,and it will improve the credibility of structure strategy.Experiment results show that this algorithm efficiently improves the precision while maintaining recall.