摘要
首先介绍粗糙集属性约简的概念并指出属性约简算法的优缺点 ,在此基础上 ,提出了 ε - 近似属性约简的概念和相应的计算方法 ,它可以比单纯的属性约简更有效地对冗余信息表进行浓缩并获得简洁的近似决策规则 ,并以一个具体的实例对此进行了分析和比较。
In this paper, the concept of attribute reduction and its functionality in the theory of rough sets is analyzed first. Then a concept of ε-approximate attribute reduction is introduced and an algorithm for findingan ε-approximate reduct is presented. It turns out that using the ε-approximate attribute reduction instead of the traditional attribute reduction may help to enhance the efficiency of data enrichment and at the same time to extract simpler and more concise approximate decision rules from a given information table. At the end of the paper, a practical example is given to show the advantages of the ε-approximate attribute reduction.
出处
《福建工程学院学报》
CAS
2003年第1期15-17,共3页
Journal of Fujian University of Technology
基金
教育部重点科研项目 ( 0 0 185 JA0 0 144 )
