中文文本分类方法研究  

Chinese Text Classification Method Research

作　　者：李凯 LI Kai(Department of Computer Technology, Anhui University of Science and Technology, Huainan 232001, China)

机构地区：[1]安徽理工大学计算机科学与工程学院,安徽淮南232001

出　　处：《电脑知识与技术》2019年第2期242-244,共3页Computer Knowledge and Technology

摘　　要：由于现实生活中大多数信息被存储为文本,因此文本挖掘具有在商业上的高潜在价值。实际应用中可以从许多信息来源中挖掘知识,然而,非结构化文本仍然是最容易获得的知识来源。该文介绍了文本分类的过程以及对三种分类器的概述,并在最后对三种分类器分别实验,以及对实验结果分析得知本实验环境下支持向量机分类器的分类效果要好于另外两种分类器。Since most of the information in real life is stored as text, text mining has high commercial potential. In practice, knowl. edge can be mined from many sources of information, however, unstructured text is still the most accessible source of knowledge. This paper introduces the process of text classification and the overview of the three classifiers. Finally, the experiments of the three classifiers and the analysis of the experimental results show that the classification effect of the support vector machine classifi. er is better than the other two classifiers.

关 键 词：文本分类 文本表示 特征选择 权重 文本分类器 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]