| 一种基于粗集的概念学习新方法 |
| |
| 引用本文： | 苏宏升,赵峰.一种基于粗集的概念学习新方法[J].重庆邮电大学学报(自然科学版),2007,19(4):405-408. |
| |
| 作者姓名： | 苏宏升 赵峰 |
| |
| 作者单位： | 兰州交通大学,自动化与电气工程学院,兰州,730070 |
| |
| 基金项目： | 甘肃省自然科学基金
,
兰州交通大学青兰工程基金 |
| |
| 摘 要： | 针对传统粗集决策方法不能对未见实例提供很好分类的缺陷，提出了一种基于粗集的概念学习新方法。
利用概念之间的偏序结构，在整个变型空间搜索与训练样例一致的假设，并通过投票的方式得到新实例最可能的
假设。仿真结果表明该方法泛化能力强，是一种有效的概念学习方法。
|
| 关 键 词： | 粗糙集 概念学习 变型空间 偏序结构 |
| 文章编号： | 1673-825X（2007）04-0405-04 |
| 收稿时间： | 2007-05-11 |
| 修稿时间： | 2007年5月11日 |
A new concept learning method based on rough set |
| |
| SU Hong-sheng,ZHAO Feng.A new concept learning method based on rough set[J].Journal of Chongqing University of Posts and Telecommunications,2007,19(4):405-408. |
| |
| Authors: | SU Hong-sheng ZHAO Feng |
| |
| Institution: | Sehool of Automatic and Electrieal Engineering, Lanzhou Jiaotong University, Lanzhou 730070, P. R, China |
| |
| Abstract: | Since the decision making methods of conventional rough set can' t give the approving classification to new instance, a new concept learning method has been proposed based on rough set theory. The method finds the hypotheses consistent with training samples by partial order structure between the concepts in the overall version space, and gets the most likely hypothesis for a new instance by voting means. The simulation result indicates that the proposed method is an effective concept learning method that possesses stronger generalized capability. |
| |
| Keywords: | rough set concept learning~ version space partial order structure |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文 |
