摘要
阐述了粗糙集理论和信息熵的概念,在此基础上提出了一种基于信息熵的属性约简算法。该算法从相对核的角度出发,将信息熵、条件信息熵和属性的重要度结合运用,优化了算法的结构,同时加快了决策表的运行速度。用CTR和Wine数据集对提出的算法进行了实验验证。结果表明,该算法能获得决策系统的最优属性约简,同时加快了运行速度。
Attribute reduction is one of the important issues of rough set. It can remove superfluous knowledge from decision system which the ability of classification is preserved, and improve efficiency of decision system. This paper expounds the basic conceptions of rough set theory and information entropy, and an algorithm of attribute reduction based on rough set and information entropy is put forward. The algorithm from the point of view of relative core, make information entropy and conditional entropy and the important degree of attributes together to use. This algorithm optimizes the structure of the original algorithm and speeds up the speed of attribute reduction. The proposed algorithm is validated by experiment through using CTR and Wine gorithm can obtain the optimal attribute reduction data set. The experimental results show that the algorithm can obtain the optimal attribute reduction of decision system, and speed up the speed.
出处
《重庆理工大学学报（自然科学）》
CAS
2013年第1期42-46,共5页
Journal of Chongqing University of Technology：Natural Science
基金
重庆市科委科技攻关项目(2008CC22)
关键词
粗糙集
信息熵
决策系统
属性约简
rough set
information entropy
decision system
attribute reduction
