| 一种基于分辨矩阵的新的属性约简算法 |
| |
| 引用本文： | 汪小燕,杨思春.一种基于分辨矩阵的新的属性约简算法[J].微机发展,2008,18(2):77-79. |
| |
| 作者姓名： | 汪小燕 杨思春 |
| |
| 作者单位： | 安徽工业大学计算机学院 安徽马鞍山243002 |
| |
| 基金项目： | 安徽省自然科学基金项目(KJ20078245) |
| |
| 摘 要： | 属性约简是粗糙集理论中的重要研究内容之一．但属性约简是一个NP难题，需要通过启发式知识实四。文中提出利用分辨矩阵求不同的条件属性组合相对于决策属性的正域的方法，并给出新的求核属性的方法。在此基础上，提出了一种利用分辨矩阵实现属性约简的新算法，该算法能快速求最少属性且实现简单，并实现了属性约简与规则提取的同步．最后通过实例证明了其正确性。
|
| 关 键 词： | 粗糙集 分辨矩阵 属性约简 决策树 |
| 文章编号： | 1673-629X(2008)02-0077-03 |
| 修稿时间： | 2007年5月24日 |
A New Algorithm for Attribute Reduction Based on Discernible Matrix |
| |
| WANG Xiao-yan,YANG Si-chun.A New Algorithm for Attribute Reduction Based on Discernible Matrix[J].Microcomputer Development,2008,18(2):77-79. |
| |
| Authors: | WANG Xiao-yan YANG Si-chun |
| |
| Affiliation: | WANG Xiao-yan, YANG Si-chun (School of Computer Science, Anhui University of Technology, Ma＇ anshan 243002,China) |
| |
| Abstract: | Attribution reduction is one of the important topics in the research on rough set theory.But the attribution reduction is a NP problem,and it needs to be realized by knowledge of elicitation method.Proposes the method that can calculate the positive region of different condition attribution combination relatively decision attribution.A new approach to compute the core attribute is also put forward in this paper.On the basic of this,the new algorithm for attribute reduction based on discernible matrix is proposed.The algorithm can get the smallest attributes quickly and be realized easily.It is realized that attribute reduction is step with rules collection.And it is proved to be workable in the practice. |
| |
| Keywords: | rough set discernible matrix attribute reduction decision tree |
| 本文献已被 CNKI 维普 等数据库收录！ |
