Abstract: The traditional ontology matching methods that use the ontology's structure to find the matches do not really make good use of the ontology's structural feature, which leads to considerable computation redundancies during the entire matching process. Therefore, a modified method named TARA was proposed to improve the matching process in this paper. The method firstly casted matching process by strictly using the ontology's structural information, and then a re-match process was applied to overcome the inevitable defect that caused by the matching process before. The experimental results show that the method has good performances in both recall and precision.

Key words: semantic Web, ontology matching, edit distance, Ontology Alignment Evaluation Initiative (OAEI), Jena

摘要： 传统的利用本体结构信息对本体做匹配的方法，并未充分利用本体的树形结构特点，致使整个本体匹配的匹配过程具有大量的冗余计算。因此，提出一种改进的基于本体树形结构的本体匹配算法TARA。该方法首先严格地以本体的树形结构为依据进行本体匹配，然后通过二次匹配来解决由于严格按照树形结构进行匹配而产生的不可避免的不足。实验结果表明，TARA方法的查全率和准确率都有较好的表现。

关键词: 语义网, 本体匹配, 编辑距离, 本体匹配竞赛, Jena