基于模糊描述逻辑的CBR的事例相似性  

On case similarity metric for CBR based on fuzzy description logic

作　　者：孙晋永[1] 钱俊彦[1] 唐成华[1] 马林威 SUN Jinyong QIAN Junyan TANG Chenghua MA Unwei(Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin 541004, China)

机构地区：[1]桂林电子科技大学广西可信软件重点实验室,广西桂林541004

出　　处：《济南大学学报（自然科学版）》2016年第5期353-363,共11页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金(61262030;61363030;61462020);桂林电子科技大学创新团队;广西高等学校高水平创新团队及卓越学者计划;广西可信软件重点实验室研究课题(KX201412)

摘　　要：针对用于基于事例推理(CBR)的描述逻辑不能表示和推理模糊语义知识的局限问题,将模糊描述逻辑F-SHIQ(D)引入CBR中,定义F-SHIQ(D)范式,据此表示包含模糊语义知识的CBR事例,提出同时考虑事例描述概念和隶属度相似性的事例相似性方法,并与现有基于描述逻辑的CBR进行对比。结果表明,在包含模糊语义信息的领域,基于F-SHIQ(D)的CBR能提升事例表示和事例相似性度量的准确性,有助于提高事例检索效率。Focused on the limitation that the existing description logics used in CBR cannot represent and reason fuzzy semantic knowledge, the fuzzy description logic F - SHIQ （D） was introduced. A normal form of F - SHIQ （D） was defined with which CBR cases involving fuzzy semantic knowledge were represented. A case＇ s similarity metric was put forward by taking similarities of both case representing concepts and case memberships into account. Compared with the existing CBR based on description logics, the result shows that, in domains with fuzzy semantic information, CBR based on fuzzy description logic improves the accuracy of case representation and case similarity metric and is great helpful in improving the efficiency of case retrieval.

关 键 词：基于事例推理 模糊描述逻辑F-SHIQ(D) 事例表示 相似性度量 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]