计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (2): 21-23.DOI: 10.3778/j.issn.1002-8331.2009.02.006

• 博士论坛 • 上一篇    下一篇

基于系统熵属性约简的差别矩阵方法

王熊彬1,郑雪峰1,徐章艳2   

  1. 1.北京科技大学 信息工程学院,北京 100083
    2.广西师范大学 计算机系,广西 桂林 541004
  • 收稿日期:2008-09-01 修回日期:2008-10-20 出版日期:2009-01-11 发布日期:2009-01-11
  • 通讯作者: 王熊彬

Method of discernibility matrix for attribute reduction based on system entropy

WANG Xiong-bin1,ZHENG Xue-feng1,XU Zhang-yan2   

  1. 1.School of Information Engineering,University of Science and Technology Beijing,Beijing 100083,China
    2.Department of Computer,Guangxi Normal University,Guilin,Guangxi 541004,China
  • Received:2008-09-01 Revised:2008-10-20 Online:2009-01-11 Published:2009-01-11
  • Contact: WANG Xiong-bin

摘要: 在已有的相对属性约简算法中,差别矩阵方法是一种设计属性约简的常用方法。基于系统熵的属性约简是一种新型的属性约简,对于这种属性约简,目前还没有差别矩阵方法去设计其属性约简算法。为此,首先构造了一种新的差别矩阵,同时给出了该差别矩阵的属性约简定义,然后证明了基于新差别矩阵的属性约简与基于系统熵的属性约简是等价的。最后用新差别矩阵设计了一个基于系统熵属性算法,并用实例说明了新算法。

关键词: 粗糙集, 属性约简, 系统熵, 差别矩阵, 算法复杂度

Abstract: The method of discernibility matrix is a kind of general method for designing the algorithm of attribute reduction in the existed relative attribute reduction.Attribute reduction based on system entropy is the new attribute reduction.To this attribute reduction,there has no the method of discernibility matrix for designing the algorithm of attribute reduction.So a new discernibility matrix is designed.At the same time,the definition of attribute reduction based on this new discernibility matrix is proposed.Then it is proved that attribute reduction based on the new discernibility matrix is equal to that based on system entropy.At last,a new algorithm of attribute reduction based on system entropy with the new discernibility matrix is provided.And an example is used to illustrate the new algorithm.

Key words: rough set, attribute reduction, system entropy, discernibility matrix, algorithm complexity