中文摘要：

针对当前在大规模本体映射方面存在的不足,提出一种新的基于遗传算法的大规模本体分块与映射方法.首先,对本体进行预处理,把本体表示成有向无环图,将本体分块问题转换成图分割问题.然后采用基于遗传算法的GPO（Genetic-Parti-tion-Ontology）算法对有向无环图进行分块,该算法不需要输入分块个数;最后利用基于本体块结构和基于参考点相结合方法找到正确的块映射.理论分析和实验结果表明,本文提出的映射方法能取得很好的结果.

英文摘要：

Due to the shortcomings in current large scale ontology mapping,proposesd a new method for large ontology partition and mapping based on genetic algorithm.Firstly,the ontologys are represented in DAG（Directed Acyclic Graph）structures by preprocessing.So,the ontology partition question is turned into a graph partition problem.Then we use GPO（Genetic-Partition-Ontology） algorithm which based on genetic algorithm to partition the DGA,and the algorithm don′t need the parameter of block number.Finaly,blocks from different ontology are matched by combining two methods of ontology blocks structure as well as predifind anchors.Theoretical analysis and experimental results shows that the method proposed by this paper can acquire good matching results.