中文摘要：

针对已有的基于进化算法的本体匹配技术要求事先提供完整的标准匹配结果的缺陷,提出一种基于局部标准匹配结果的本体匹配技术.首先提出通过本体概念聚类算法以构建局部标准匹配结果,并在此基础上建立基于局部标准匹配结果的本体匹配问题的多目标优化模型;然后设计一种基于局部标准匹配结果的MOEA/D算法;最后,采用2012年的本体匹配评价竞赛的测试数据集进行实验,实验结果表明所提出的方法是有效的.

英文摘要：

To overcome the drawback that an entire standard ontology alignment must be provided beforehand in current evolutionary-algorithm-based ontology matching technologies,we propose an ontology matching approachbased on local standard ontology alignment. First,we utilize the ontology concept of the clustering algorithm to build the local standard ontology alignment,and then construct a multi-object optimization model for the ontology matching problem based on the local standard ontology alignment. Then,we propose a multiobjective evolutionary algorithm based on decomposition（ MOEA / D） that is also based on the local standard ontology alignment. We conducted an experiment using a testing dataset from the ontology alignment evaluation initiative 2012,and the results demonstrate the effectiveness of the proposed approach.