| 基于评价指数的属性约简算法 |
| |
| 引用本文： | 陈鑫影,黄明,李瑞.基于评价指数的属性约简算法[J].大连铁道学院学报,2006,27(2):51-54. |
| |
| 作者姓名： | 陈鑫影 黄明 李瑞 |
| |
| 作者单位： | 大连交通大学软件学院 辽宁大连116028 |
| |
| 摘 要： | 从粗糙集理论出发,提出可分辨关系和对象差异矩阵的概念,并以此为基础构造出基于评价指数的属性约简算法.算法采用自底向上的方法,利用属性的评价指数作为启发式信息,求得属性集合的约简.在多个UC I数据集上进行试验,理论分析和实验数据表明,文中提出的属性约简算法是有效、可行的,具有较高的效率.
|
| 关 键 词： | 数据挖掘 粗糙集 属性约简 |
| 文章编号： | 1000-1670（2006）02-0051-04 |
| 收稿时间： | 2005-12-05 |
A Heuristic Algorithm of Attribute Reduction Based on Evaluation Index |
| |
| CHEN Xin-ying, HUANG Ming, LI Rui.A Heuristic Algorithm of Attribute Reduction Based on Evaluation Index[J].Journal of Dalian Railway Institute,2006,27(2):51-54. |
| |
| Authors: | CHEN Xin-ying HUANG Ming LI Rui |
| |
| Institution: | School of Software Technology, Dalian Jiao Tong University, Dalian 116028, China |
| |
| Abstract: | The concepts of distinguishable relation and distinguishable matrix are defined based on rough set theory. Against the new concepts, a new algorithm of heuristic algorithm of attribute reduction based on evaluation index is also propose. The heuristic algorithm adopts the bottom-up design, and achieves attribute reduction with the heuristic information-evaluation index. Theoretical analysis and experiment data show that the algorithm is feasible and accuracte and efficient. |
| |
| Keywords: | data mining rough set attribute reduction |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
