| 改进可辨识矩阵的启发式知识约简及规则算法 |
| |
| 引用本文： | 金贵林,崔炳谋.改进可辨识矩阵的启发式知识约简及规则算法[J].兰州铁道学院学报,2005,24(4):92-95. |
| |
| 作者姓名： | 金贵林 崔炳谋 |
| |
| 作者单位： | 兰州交通大学交通运输学院,甘肃兰州730070 |
| |
| 摘 要： | 首先提出了改进可辨识矩阵,然后对属性重要性进行了重新定义,并且将模糊聚类应用于属性重要性划分,从而在此基础上提出了一种启发式知识约简及决策规则提取算法.给出了新的不相容率的概念,用于描述知识数据库中数据不相容特征和修正提取的决策规则的决策错误率.经过实验,该算法具有优良的性能.
|
| 关 键 词： | 改进可辨识矩阵 粗糙集 知识约简 决策规则 属性重要性 不相容率 模糊聚类 |
| 文章编号： | 1001-4373(2005)04-0092-04 |
| 收稿时间： | 2004-12-07 |
| 修稿时间： | 2004年12月7日 |
A Heuristic Algorithm for Reduction of Knowledge and Abstracting Decision Rules Based on the Improved Discernibility Matrix |
| |
| Jin Guilin,Cui Bingmou.A Heuristic Algorithm for Reduction of Knowledge and Abstracting Decision Rules Based on the Improved Discernibility Matrix[J].Journal of Lanzhou Railway University,2005,24(4):92-95. |
| |
| Authors: | Jin Guilin Cui Bingmou |
| |
| Abstract: | The improved discernibility matrix is put forward, the attribute importance is redefined and the Fuzzy Cluster is applied for the classification of attribute importance, then based on which a heuristic algorithm for reduction of knowledge and abstracting decision rules is proposed. A new conception of rate of incompatible is also put forward,and it is used to describe the property of incompatible of data in knowledge database and to correct the rate of error decision of decision rules. The experimental results show that this algorithm has excellent quality. |
| |
| Keywords: | improved discernibility matrix rough set reduction of knowledge decision rules attribute significance rate of incompatible fuzzy cluster |
| 本文献已被 维普 万方数据 等数据库收录！ |
