Abstract: In research of ontology mapping, the mapping between real ontology or large-scale ontology is difficult. This paper proposes a mapping method for large class hierarchies, which is one kind of the large-scale ontologies. Based on ontology structure and semantic distance between classes, the classes can be put in a multi-dimension space by using Vector Space Model(VSM). Two large class hierarchies are partitioned into small blocks respectively by applying CURD clustering algorithm, and then blocks from different hierarchies are matched by comparing the reference points. Experimental results show that the method performs well on the test cases.

Key words: large-scale ontology, clustering, block, mapping

摘要： 在本体的映射研究中，现实本体或大本体之间的映射算法是研究的难点。该文提出一种针对大层次本体的映射方法。根据本体的结构和概念之间的语义距离，应用向量空间模式(VSM)将概念表示成多维空间中的点。在此基础上，应用聚类算法(CURD)对概念进行聚类，形成若干个语义上相对独立的块，在2个本体的块之间根据参考点建立映射关系。实验结果表明，该方法在测试数据集上能得到较好的映射结果

关键词: 大本体, 聚类, 块, 映射

CLC Number: 

• TP301