Abstract: The paper affords an improved method to solve the shortage of the traditional similarity in ontology mapping.First according to the characteristic of ontology tree and the idea of the data digging，the paper affords an advanced measure of setting candidate mapping set，in order to reduce the runtime of the account.Second according to the characteristic of the ontology and concept，the paper designs an improved method of similarity，synthetizing the concept name，instance，attribute，structure and relation.The method meliorates the problem，such as unilateralism and faultiness，in the traditional calculational methods，and improves the precision and recall in the ontology mapping.Experiment indicates that the method is better than the Glue at Runtime Complexity、Precision and Recall.

摘要： 针对本体映射中概念相似度计算中存在的一些不足之处，提出了一种改进的方法。首先根据本体概念树的特点以及数据挖掘的思想，提出了一种改进的侯选映射集设置的方法，减少了相似度计算的工作量。进而根据本体和本体中概念的特点，综合概念名称、概念实例、概念的属性、结构、关系等因素。设计了一种改进的相似度的计算方法。改善了相似度计算中存在的片面性和不完善性问题，提高了本体映射的查全率和查准率。初步实验表明该算法在计算复杂度、查全率和查准率上都要优于Glue方法。