纯Peer-to-Peer环境下有效的本体匹配方法

Efficient Ontology Matching Processing in Pure Peer-to-Peer Network

目前大多数的Peer-to-Peer(P2P)系统由于欠缺对语义的支持而不能很好地应用于知识共享领域.可能的方法是采用本体(ontology)来描述Peer节点共享资源的语义.针对纯P2P环境下不同节点单独维护局部本体异构的特点,提出了基于历史查询记录的纯P2P环境下的本体匹配方法HOM.首先利用历史查询记录挖掘出概念之间可能存在的映射关系,然后利用向量空间模型VSM(vector space model)方法来识别概念在语言学上可能存在的联系,进而修正概念之间的映射关系,建立不同节点间本体概念的匹配.分析和仿真表明,与现有方法相比,HOM方法考虑了节点的自治特性,并提高了本体匹配的精度.

Most of the existing peer-to-peer（P2P） systems suffer from the lack of the support for semantic contents,thus making them unsuitable for knowledge sharing.To solve the problem,a prospective method is suggested,introducing ontology to describe the resources to be shared in P2P systems semantically.In a pure P2P environment,where the different peers are keeping up their characteristics of locally ontological heterogeneity individually,an HOM（history-based ontology matching） method is proposed on the basis of historical query logs.Mining the possible mapping relationship between different ontological concepts from historical query logs,the proposed method applies VSM to the recognition of the possible linguistic relationship between ontological concepts.Then,the mapping relationship between concepts is corrected to match the ontological concepts between different peers.Analytic and experimental results show that,compared with the existing methods,the HOM method considers the autonomy of peers and is more precise on matching different ontology.