| 文本分类中一种改进的特征选择方法 |
| |
| 引用本文： | 刘海峰,王元元,张学仁.文本分类中一种改进的特征选择方法[J].情报科学,2007,25(10):1534-1537. |
| |
| 作者姓名： | 刘海峰 王元元 张学仁 |
| |
| 作者单位： | 1. 解放军理工大学,指挥自动化学院,江苏,南京,210007
2. 解放军理工大学,理学院,江苏,南京,210007 |
| |
| 摘 要： | 本文提出了一种改进的基于互信息的特征选择方法，与改进的TF-IDF权值公式相结合对文本特征进行选择，提高了特征项信息利用效率。试验表明，该算法提高了文本分类正确率。
|
| 关 键 词： | 信息增益 互信息 信息比值 特征选择 文本分类 |
| 文章编号： | 1007-7634（2007）10-1534-04 |
| 修稿时间： | 2007-03-29 |
An Improved Feature Selection Method in Text Classification |
| |
| LIU Hai-feng,WANG Yuan-yuan,ZHANG Xue-ren.An Improved Feature Selection Method in Text Classification[J].Information Science,2007,25(10):1534-1537. |
| |
| Authors: | LIU Hai-feng WANG Yuan-yuan ZHANG Xue-ren |
| |
| Institution: | 1. Institute of Command Automation, PLAUST, Nanfing 210007, China ; 2. Institute of Sciences, PLAUST, Nanfing 210007, China |
| |
| Abstract: | In this paper, we put forward an improved feature selection method that based on MI. Combined with the improved weighting formula TF - IDF to select text feature, we have increased the using efficiency about the character information. The experiment shows that this method has improved the precision of the text classification. |
| |
| Keywords: | information gain multi - information information ratio feature selection text classification |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
