一种改进的粗糙集属性约简启发式算法  

A modified heuristic algorithm of attribute reduction in rough set

作　　者：索永强[1] 马力[2] 谭薇[1] 

机构地区：[1]西安邮电学院计算机科学与技术系,陕西西安710121 [2]西安邮电学院信息中心,陕西西安710121

出　　处：《西安邮电学院学报》2009年第5期116-120,共5页Journal of Xi'an Institute of Posts and Telecommunications

摘　　要：作为数据挖掘的重要工具,粗糙集理论被广泛的应用于关系数据库中属性相关性描述、属性集约简、属性重要性度量、规则发现等方面。该文在分析基于信息系统的粗糙集理论的基础上,对基于分辨矩阵的属性约简算法进行了详尽的描述。针对该算法存在的时间和空间性能不理想问题,提出度量单个条件属性对系统概念贡献程度的关联度的概念,以此作为启发式信息对原算法进行改进,得到条件属性的约简。理论分析及实验结果表明该算法具有较好的约简效果及更高的运行效率,为粗糙集理论更广泛地应用于具体的实践提供了一种方法。As a useful tool for data mining, the rough set theory is widely used in the description of the correlation between attributes of relational database, the reduction of the attribute set, the counting of an attribute importance compared to other attribute importance, the discovery of rules, and so on. This paper discusses the attribute reduction in rough set theory. First, on the basis of analyzing the rough set theory, a detailed description of attribute reduction algorithm based on the discernible matrix is given. Second, as the traditional algorithm has relatively poor efficiency in both time and space when obtaining attribute reduction, the relationship degree, which describes the contribution of one condition attributes to decision attribute, is introduced into rough set to value the importance of attribute. And then a modified algorithm is given by using the relationship degree as heuristic information, Theory analysis and the experimental results show this algorithm costs less time and reduces the number of condition attributes than other algorithms. It establishes a method by which rough set theory is widely used in concrete practice.

关 键 词：粗集理论 属性约简 关联度 启发式算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]