Rough集理论代数观与信息观的差异量化分析  

Quantitative Analysis of the Difference Between the Algebra View and Information View of Rough Set Theory

作　　者：王国胤[1] 安久江[1] 吴渝[1] 

机构地区：[1]重庆邮电学院计算机科学与技术研究所,重庆400065

出　　处：《小型微型计算机系统》2005年第7期1187-1190,共4页Journal of Chinese Computer Systems

基　　金：新世纪优秀人才支持计划资助;国家自然科学基金(60373111)资助;重庆市教委科学技术研究项目(040505)资助

摘　　要：决策表是Rough集理论的处理对象,其核属性的计算往往是信息约简过程的出发点和关键.代数观和信息观是Rough集理论研究中的两种主要理论观点和方法.本文将针对决策表核属性的计算问题,探讨Rough集理论代数观和信息观这两种形式的关系,通过仿真实验,得到它们在决策表核属性问题上的统计量化差异,并发现在包括大量不相容信息的决策表系统中这两种观点之间的差异将达到极端情况.The attribute core of a decision table is often the start point and key of many decision information system reduction procedures based on rough set theory. The algebra and information vienpoints are two main methods of rough set theory. In this paper, based on the problem of calculating the attribute core of a decision table, we study the relationship between the algebra viewpoint and information viewpoint. Through simulation experiment, we find the quantitative difference of the attribute cores of a decision table from these two viewpoints of rough set theory. Especially, we find that their difference will be maximized in decision tables containing much inconsistent information.

关 键 词：ROUGH集 决策表 核属性 代数观 信息观 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]