结合粗糙集理论与扩张矩阵理论的数据挖掘方法  

A data mining approach with combination of rough set theory and expansion matrix theory

作　　者：赵士亮[1] 叶东毅[1] 陈云化[1] 

机构地区：[1]福州大学信息科学与技术学院,福建福州350002

出　　处：《福州大学学报（自然科学版）》2001年第4期49-52,共4页Journal of Fuzhou University(Natural Science Edition)

基　　金：教育部重点科研项目 (0 0 185 ) ;福建省自然科学基金资助项目 (A0 0 10 0 0 9) ;福建省教育厅科研项目(JA0 0 144 )

摘　　要：提出将粗糙集理论、扩张矩阵理论进行有机结合的新方法 ,该方法吸收了两者的优点同时消除了两者的缺点 .实践证明 ,该方法可以十分有效地从数据库中挖掘出准确而精悍的知识 .Rough Set Theory has an inherent ability of dealing with imprecise and incomplete information while in most cases it produces a large quantity of rules, thus reducing readability. On the contrary, expansion matrix theory canmerely handle precise and complete information while it is efficient to get formulas that cover as many positive examples as possible yet meanwhile kick away all negative ones. So we present a new data mining approach that combines rough set theory and expansion matrix theory in order to absorb merits of both while eliminate drawbacks of both. And we find it quite efficient to discover concise knowledge in databases.

关 键 词：粗糙集理论 扩张矩阵理论 数据库 数据挖掘 人工智能 分类规则 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP311.13[自动化与计算机技术—控制科学与工程]