Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (22): 171-174.

Previous Articles     Next Articles

Method of attribute weight for incomplete information system

TAN Zongfeng1, XU Zhangyan2, LI Xiaoyu2   

  1. 1.College of Mathematical Sciences, Guangxi Normal University, Guilin, Guangxi 541004, China
    2.College of Computer Science & Information Engineering,Guangxi Normal University, Guilin, Guangxi 541004, China
  • Online:2012-08-01 Published:2012-08-06

不完备信息系统的一种权重确定方法

谭宗凤1,徐章艳2,李晓瑜2   

  1. 1.广西师范大学 数学科学学院,广西 桂林 541004
    2.广西师范大学 计算机科学与信息工程学院,广西 桂林 541004

Abstract: This paper deeply studies the attribute importance represented by rough set. Considering the drawbacks of the existing methods of attribute weight assignment based on rough set and combining the overall and individual significance degree of each attribute in the conditional attributes, this paper proposes a method of attribute weight assignment for incomplete information system, and analyzes the rationality. A case proves that improved weighting method can solve the existing problem.

Key words: weight, incomplete information system, attribute significance degree, condition entropy

摘要: 深入分析了粗糙集表征的属性重要度,针对现有粗糙集理论的属性权重确立方法的不足,结合属性集中条件属性的整体重要度和系统中条件属性的个体重要度,提出了在不完备信息系统的一种权重确定方法,分析了其合理性。通过算例分析表明,新的不完备信息系统的权重确定方法可以解决原有粗糙集权重确定方法存在的问题。

关键词: 权重, 不完备信息系统, 属性重要度, 条件熵