计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (29): 63-67.

• 学术探讨 • 上一篇    下一篇

一个基于语义模块的交互式本体匹配框架

文敦伟1,2,樊小虎1   

  1. 1.中南大学 信息工程与技术学院,长沙 410083
    2.阿萨巴斯卡大学 计算机与信息系统学院 阿萨巴斯卡 T9S3A3 加拿大
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-10-11 发布日期:2007-10-11
  • 通讯作者: 文敦伟

Semi-auto framework for module based ontology matching

WEN Dun-wei1,2,FAN Xiao-hu1   

  1. 1.School of Information Science and Engineering,Central South University,Changsha 410083,China 2.School of Computing and Information Systems,Athabasca University,Athabasca T9S3A3,Canada
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-10-11 Published:2007-10-11
  • Contact: WEN Dun-wei

摘要: 本体匹配是用来解决异质本体间互操作问题的一种技术手段。目前,大多数关于本体匹配的研究都集中在了如何提高匹配结果的质量上。然而,一方面,在许多情况下,匹配结果的正确与否直接依赖于用户的判断,另一方面,由于一些描述现实世界的本体十分庞大,匹配工具往往不能及时为用户提供可供确认的匹配对。为此,提出了一种基于语义模块的交互式本体匹配框架。借助信息论的相关知识,将本体聚类成语义模块。用户利用模块核心结点信息对模块的内容进行推断,从而将大规模的本体匹配任务转换为数个规模较小的语义模块间的匹配任务。通过合理地增大用户在匹配过程中的作用,试图在保证匹配质量的同时提高匹配效率。已获得的实验结果表明该方法能显著提高本体匹配任务的效率。

关键词: 本体匹配, 聚类算法, 信息论

Abstract: Ontology matching aims at dealing with interoperation problems between heterogeneous ontologies.Recently,many researches are focused on improving the matching tool’s quality.However,in one hand,the goals of matching depend heavily on the intension of the users;in the other hand,for a matching task with large ontologies,it is very difficult to efficiently present the matching results to a human engineer to validate.Hence,in this paper,we have proposed a semi-auto framework for semantic-modular based ontology matching.Addressed with information theory,the approaches employ a clustering algorithm to modularize ontologies.With the information of the semantic core,user can predicate the content of the clusters,thus transform the large ontology matching task into several smaller ones.By amplifying user’s effects wisely,we tried to achieve high efficiency while keeping the match qualities.The results have obtained so far show that this measure could improve the efficiency significantly.

Key words: ontology matching, clustering algorithm, information theory