| 模糊kNN在文本分类中的应用研究 |
| |
| 引用本文： | 林永民,朱卫东.模糊kNN在文本分类中的应用研究[J].计算机应用与软件,2008,25(9). |
| |
| 作者姓名： | 林永民 朱卫东 |
| |
| 作者单位： | 1. 河北理工大学经济管理学院,河北,唐山,063009
2. 北京交通大学计算机与信息技术学院,北京,100044 |
| |
| 摘 要： | 自动文本分类是根据已经分配好类标签的训练文档集,来对新文档分配类标签.针对模糊kNN算法用于文本分类的性能进行了一系列的实验研究与分析.在中英文两个不同的语料集上,采用四种著名的文本特征选择方法进行特征选择,对改进的模糊kNN方法与经典kNN及目前广泛使用的基于相似度加权的kNN方法进行实验比较.结果表明,在不同的特征选择方法下,该算法均能削弱训练样本分布的不均匀性对分类性能的影响,提高分类精度,并且在一定程度上降低对k值的敏感性.
|
| 关 键 词： | 文本分类 模糊kNN 特征选择 |
STUDY ON THE APPLICATION OF FUZZY kNN TO TEXT CATEORIZATION |
| |
| Lin Yongmin,Zhu Weidong.STUDY ON THE APPLICATION OF FUZZY kNN TO TEXT CATEORIZATION[J].Computer Applications and Software,2008,25(9). |
| |
| Authors: | Lin Yongmin Zhu Weidong |
| |
| Affiliation: | Lin Yongmin1 Zhu Weidong2 1(College of Economics , Management,Hebei Polytechnic University,Tangshan 063009,Hebei,China) 2(School of Computer , Information Technology,Beijing Jiaotong University,Beijing 100044,China) |
| |
| Abstract: | Automated text categorization is defined as the process of assigning category labels to new documents based on the likelihood suggested by a training set of labelled documents.A study on the performance of fuzzy kNN for text categorization is presented in two different language corpus of Chinese and English.Four famous approaches to text feature selection are adopted.The improved fuzzy kNN method is compared to the conventional kNN method and the kNN method based on similarity-weighting,and the experimental... |
| |
| Keywords: | Text categorization Fuzzy kNN Feature selection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
