一种基于本体的概念相似度计算及其应用

概念的语义相似度研究,是知识表示以及信息检索领域中的一个重要内容。本文提出了基于语义相似度和相关度的综合概念相似度计算方法,考虑了语义距离和本体库特征,加入概念的信息重合度、概念的深度、概念的密度和不对称因子的辅助影响。通过实验和两种传统的语义相似度计算方法进行对比,本方法能更好地区分本体树中不同关系的概念对,验证了该方法的有效性。

A Concept Similarity Computation Based on Ontology and its Application

Semantic similarity of concepts is an important content in the field of knowledge representation and information retrieval. This paper puts forward an integrated method based on semantic similarity and semantic relevance,we think about the semantic distance and the characteristics of ontology,the concept’s information coincidence degree,the depth of concept,the density of concept and symmetry factor.Compared with two traditional semantic similarity computation methods,this method can better distinguish the different concept in ontology tree and this method is effective.