| 文本分类的几种方法研究 |
| |
| 引用本文： | 沙俐敏.文本分类的几种方法研究[J].南方冶金学院学报,2004,25(1):50-54. |
| |
| 作者姓名： | 沙俐敏 |
| |
| 作者单位： | 上海第二工业大学 上海201209 |
| |
| 摘 要： | 经过训练和统计对每一类文本形成特征的权重向量,利用K-最近距离的方法对测试集进行分类.Sleepingexpert算法采用正权重和负权重较好地描述了多义词的特性,该文在原算法中插入了一种权重补偿模块,其目标是实现权重和当前概念的一致性,具有更好的分类性能.
|
| 关 键 词： | 文本分类 基于K-最近距离 Sleepingexpert 概念推理网 权重 |
| 文章编号： | 1007-1229(2004)01-0050-05 |
| 修稿时间： | 2003年11月2日 |
Several Methods for Text Classification |
| |
| SHA,Li-min.Several Methods for Text Classification[J].Journal of Southern Institute of Metallurgy,2004,25(1):50-54. |
| |
| Authors: | SHA Li-min |
| |
| Abstract: | After training, we get the vector space model of the text categorization. The classification of the input text is decided by K-nearest-neighbor.The Sleeping expert algorithm characterizes polysemy with positive and negative weights. This paper uses a compensation module inserted into the classical sleeping expert algorithm to modify the positive weight of |
| |
| Keywords: | based on their statistical weight and context to maintain consistency between the weight and the keywords'current concept The new algorithm can markedly improve the classification performance Key words: text classification K-nearest-neighbor |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
