一种改进的最小属性约简算法  

An Improved Algorithm on Mininum Attribute Reduction

作　　者：薛胜军[1,2] 郭强[1] 

机构地区：[1]武汉理工大学计算机科学与技术学院,武汉430063 [2]南京信息工程大学计算机与软件学院,南京210044

出　　处：《武汉理工大学学报（交通科学与工程版）》2012年第3期515-518,共4页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：国家自然科学基金项目资助(批准号:60702076)

摘　　要：最小属性约简是粗糙集理论中的一个重要内容,通过对属性约简算法的研究,指出基于U/{a}划分的最小属性约简算法中的不足,并提出了一种改进的最小属性约简算法.该算法以基于等价类的方法对论域进行压缩,有效地降低了算法的空间复杂度,用基于核的方法求解最小约简,从而保证可获得最小约简.实验结果表明,该算法具有复杂性低、存储空间小和效率高的特点,是一种有效的最小属性约简算法.Mininum attribute reduction is an important part of rough set theory. By researching attrib- ute reduction algorithms,the drawbacks of mininum attribute reduction based on U/{a} partition is derived,and an improved mininum attribute reduction algorithm is proposed in this paper. First, the improved algorithm compresses the universe by the equivalence classes algorithm , this effectively re- duces the space complexity of the improved algorithm. Then, the algorithm of the computation of a core is used to get the mininum attribute reduction,this algorithm ensure that it can get the mininum attribute reduction. Experimental results show that the improved algorithm has the lower complexity, smaller storage space and higher efficiency, and is an effective minimum attribute reduction algorithm.

关 键 词：粗糙集 最小属性约简 核 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]