基于依赖度之差的属性重要性评分  

Significance of Attribute Evaluation Based on Dependable Difference

作　　者：王小菊[1] 蒋芸[1] 李永华[1] 

机构地区：[1]西北师范大学数学与信息学院,甘肃兰州730070

出　　处：《计算机技术与发展》2009年第1期67-70,共4页Computer Technology and Development

基　　金：甘肃省自然科学基金项目(3ZS051-A25-042);西北师范大学2006-2010年度重点学科资助项目

摘　　要：目前粗糙集决策表中条件属性的重要性基本上是用条件属性的依赖度进行评判的。在决策表约简中,利用条件属性的依赖度进行评判可能会造成某些重要的条件属性的简单丢弃,影响了决策的准确性。因此提出并分析了基于依赖度之差的属性重要性的判断方法,该方法可以确保得到决策表的重要属性,得出了用依赖度之差判断属性重要性更加准确的结论,同时,给出了依赖度之差的求解步骤和算法,并通过实例验证了用依赖度之差判断属性重要性的有效性。At present basically significance of conditional attribute in the decisional table based on rough sets is evaluated through dependability of conditional attribute. Evaluation by using dependability of conditional attribute possible discard some important conditional attribute, it may affect the accuracy of the decision - making. Proposes and analyzes an approach of judging importance of attribute based on dependable difference, the approach can get important attribute, It is obtained that judging importance of attribute based on dependable difference of attribute is mere accurate. And dependable difference of attribute detailed solve procedure and arithmetic is provided. The efficiency of judging importance of attribute based on dependable difference of attribute is proven through the example.

关 键 词：粗糙集 决策表 依赖度 依赖度之差 属性重要性 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]