论文部分内容阅读
本文基于粗集理论,提出了一种新的规则提取法LBR(Learning By Rough Sets),并对LBR与另一种已有的规则提取法LEM1,即全局覆盖算法(global covering algorithm)进行了比较和讨论.基于比较的结果,得出了将LEM1改进后的LEM3.LBR不但可用于普通的决策表规则提取,更多地可应用于基于模糊划分的规则提取.LBR的提出,极大地简化和丰富了规则提取算法,在已知数据中可获取更为丰富的信息量.而LEM3的使用,则是在将“依赖”(depend on)这一概念推广的基础上,更灵活地使用“覆盖”(covering),扩大了获取规则的范围.LBR和LEM3因其各自不同的优点,在数据挖掘和智能领域均具有广泛的应用前景.
Based on rough set theory, this paper proposes a new Learning By Rough Sets (LBR) algorithm and compares LBR with another existing rule extraction method, LEM1, that is, the global covering algorithm And discussed.According to the result of the comparison, it is concluded that the LEM3 LEM with improved LEM1 can be used not only for ordinary decision table extraction, but also for rule extraction based on fuzzy partitioning.LB is greatly simplified and Enriched rules extraction algorithm, can obtain richer information in known data, and the use of LEM3 is more flexible based on the promotion of the concept of “depend on” “Covering ” (covering), expanding the scope of access rules.LBR and LEM3 because of their different advantages, in the field of data mining and intelligence has a wide range of applications.