| 基于信息增益的属性约简算法 |
| |
| 引用本文： | 苟光磊,黄丽丰,邹航.基于信息增益的属性约简算法[J].重庆工学院学报,2007,21(11):90-92,116. |
| |
| 作者姓名： | 苟光磊 黄丽丰 邹航 |
| |
| 作者单位： | 重庆工学院计算机科学与工程学院,重庆400050 |
| |
| 基金项目： | 重庆市教委科学技术研究项目（KJ060620）. |
| |
| 摘 要： | 以信息增益作为属性重要性的度量方法，提出了一种基于信息增益的属性约简算法．该算法总是优先考虑对于决策更为重要的属性，用条件属性对决策属性的信息增益作为条件属性的属性重要性度量，并以此度量作为启发式信息，算法从空集开始逐步将重要的属性加入到选择属性集，直到决策表达到一致分类时结束．并通过实例分析验证了该算法能有效地对属性进行约简，同时可以得到简单规则集．
|
| 关 键 词： | 粗集 属性约简 信息增益 |
| 文章编号： | 1671-0924（2007）06-0090-03 |
| 收稿时间： | 2007-04-01 |
| 修稿时间： | 2007-04-01 |
Algorithm for Attributes Reduction Based on Information Gain |
| |
| GOU Guang-lei, HUANG Li-feng, ZOU Hang.Algorithm for Attributes Reduction Based on Information Gain[J].Journal of Chongqing Institute of Technology,2007,21(11):90-92,116. |
| |
| Authors: | GOU Guang-lei HUANG Li-feng ZOU Hang |
| |
| Affiliation: | School of Computer Science and Engineering, Chongqing Institute of Technology, Chongqing 400050, China |
| |
| Abstract: | An algorithm for attributes reduction based on information gain, Which is used to measure the attributes import, is introduced in this paper. The more important attribute is taken into account in the first place. Information gain for condition attributes to decision attributes used to meaure the import is used as heursitic information. The most important attribute was added to the selected attributes set from empty set. The algrithm was terminated when consisitent classficiation was obtained in the decision table. The instance analys is shows the algorithm can effectively reduce the decision system, and the brief mhs can be obtained. |
| |
| Keywords: | rough sets attributes reduction information gain |
| 本文献已被 维普 等数据库收录！ |
|
