The similarity calculation of concept names

Abstract

With the rapid development of the semantic web, ontology has been rapidly developed. Because of the differences on the constructors and living environments, it causes that different people will get various versions, even when they are constructing the same ontology. How to solve the problem of ontology heterogeneity is a focus issue. There are many ways to solve the ontology heterogeneous issues. But the ontology mapping is the most efficient way to solve ontology heterogeneous issues. In this paper, the concept name similarity algorithm is studied deeply. The concept similarity computing algorithm is not perfect. Thus, this paper proposes an improved concept name similarity algorithm. This paper calculates the edit distance and the semantic similarity. Then the edit distance and the semantic similarity can be integrated into two concepts. Based on the information content, this paper has put forward an improved semantic similarity algorithm. Experiments results show that this algorithm can correspond to the human graded score compared to the traditional semantic similarity algorithm.