MINI：一种可减小变更影响范围的本体演化算法.doc

MINI:一种可减小变更影响范围的本体演化算法
刘晨1,2 韩燕波1 陈旺虎1,2 王建武1,2
(1 中国科学院计算技术研究所网格与服务计算研究中心,北京 100080)
(2 中科院研究生院, 北京 100049)
摘要本体演化会影响依赖本体的服务,使其重新修订和重新部署。面对同一变更需求,不同演化实现方法造成的影响范围差别很大。当前的本体演化研究主要聚焦于如何实现变更需求以及维护变更前后本体的一致性,对于如何降低演化影响范围关注甚少。本文提出了一种可以有效减小变更影响范围的本体演化算法MINI。该算法首先分析了本体实体和服务之间的依赖关系并提出了量化变更影响范围的数学公式。根据这一公式,MINI算法将本体演化过程转变为图的启发式搜索过程,通过搜索一条影响值最小的变更路径来减小本体演化的影响范围。实验结果表明,MINI算法导致的平均变更影响范围大大小于现有的本体演化算法。该算法已在某实际项目中得以应用和验证。
关键词本体演化; 本体变更; 变更路径; 本体一致性; 影响范围
中图法分类号
MINI: An Ontology Evolution Algorithm for Reducing Impact Ranges
Chen Liu1,2 Yanbo Han1 Wanghu Chen1,2 Jianwu Wang1,2
(1 Research Center for Grid and puting, Institute puting Technology, Chinese Academy of Sciences, 100080, Beijing, China)
(2 Graduate University of the Chinese Academy of Sciences,100049, Beijing, China)
Abstract Ontology evolution is apt to impact on dependant services and cause them to be redeveloped and redeployed. However, different realizing methods for the same change requirements may result in very big different impact ranges. Today, researches of ontology evolution mainly focus on how to satisfy the change requirements and maintain the consistency of an evolving ontology. Unfortunately, few of them care about how to reduce impact ranges of ontology evolution. This paper proposes an ontology evolution algorithm called MINI, which can effectively reduce impact ranges for an evolution process. Firstly, through deeply analyzing dependant relations among ontology entities and services, the MINI algorithm establishes a math formula to quantify impact ranges. Based on this formula, the MINI algorithm transforms an ontology evolution process into a heuristic graph searching process. Through searching an ontology change path which has minimal impact value, the impact range of an evolution process is greatly reduced. The experiment results show that the average impact ranges caused by the MINI algorithm is greatly less than those of other evolution
algorithms. In the context of some practical a