基于程度粗糙集模型的近似约简和近似相对约简  

Approximate Reduction and Approximate Relative Reduction Based on Degree Rough Set Model

作　　者：戴岱[1] 秦克云[1] 李众立[2] 

机构地区：[1]西南交通大学理学院应用教学系,成都610031 [2]西南科技大学,四川绵阳621010

出　　处：《西南科技大学学报》2003年第4期5-7,11,共4页Journal of Southwest University of Science and Technology

摘　　要：知识约简是在保持知识库分类能力不变的条件下,删除其中不相关或不重要的知识,它是粗糙集理论的核心内容之一。基于程度粗糙集模型提出了知识的近似约简和近似相对约简的定义,并且讨论了它们的一些相关性质。近似约简和近似相对约简是Pawlak粗糙集模型下的约简和相对约简的推广,它们能够在一定误差允许下约简更多的知识,使问题更加简化,同时也为获取近似决策规则奠定了基础。Knowledge reduction is deleting irrelevant or unimportant knowledge in the knowledge base un der the condition of keeping the classifying ability of the knowledge base; it is the core part of rough set theory. In this paper, the definition of approximate reduction and approximate relative reduction based on degree rough set model is brought forward, at the same time some of its correlative characteristics are dis cussed. Approximate reduction and approximate relative reduction is.the extending of the reduction and relative reduction based on Pawlak rough set model, which can reduce more knowledge, simplify the problem, and simultaneously on condition that a certain error is allowed, establish the foundation of obtaining approximate decision-making regulation.

关 键 词：程度粗糙集模型 近似约简 近似相对约简 知识库 知识约简 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]