| 一种改进的基尼指数特征权重算法 |
| |
| 引用本文： | 任国锋,李德华,潘莹. 一种改进的基尼指数特征权重算法[J]. 计算机与数字工程, 2010, 38(12) |
| |
| 作者姓名： | 任国锋 李德华 潘莹 |
| |
| 作者单位： | 华中科技大学图像识别与人工智能所;广西大学信息网络中心; |
| |
| 摘 要： | 文本分类中普遍应用的TF-IDF特征权重算法没有引入特征项的纯度和类别属性.在结合基尼指数原理和TF-IDF特征权重算法基础上,提出一种基于基尼指数的特征权重改进算法,在计算特征权重时引入特征项的纯度和分类的已知类别属性.进一步,设计了两种特征权重算法的对比实验,并在SVM分类器和kNN分类器下选取不同的特征项数目进行多次实验.实验结果表明,该改进的基尼指数特征权重算法有更好的效果.
|
| 关 键 词： | 改进算法 指数特征 权重算法 Gini Index Based 特征权重 基尼指数 特征项 类别属性 SVM分类器 实验结果 指数原理 文本分类 对比实验 纯度 kNN分类 设计 计算 基础 |
An Improved Algorithm for Feature Weight Based on Gini Index |
| |
| Ren Guofeng,Li Dehua,Pan Ying. An Improved Algorithm for Feature Weight Based on Gini Index[J]. Computer and Digital Engineering, 2010, 38(12) |
| |
| Authors: | Ren Guofeng Li Dehua Pan Ying |
| |
| Affiliation: | Ren Guofeng1) Li Dehua1) Pan Ying1),2)(Institute for Pattern Recognition and Artificial Intelligence,Huazhong University of Science and Technology1),Wuhan 430074)(Information Network Center,Guangxi University2),Nanning 530004) |
| |
| Abstract: | The universally used TF-IDF feature weight algorithm in the text categorization does neither introduce the purity of feature term,nor the known category property.So an improved feature weight algorithm based on the theory of gini index is proposed in the paper,which takes the purity of feature term and the known category property into account.And then experiments are designed to compare the improved algorithm with the TF-IDF algorithm,with different feature numbers in the SVM and the kNN classifier.The resu... |
| |
| Keywords: | text categorization feature weight gini index |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
