文本分类中特征选择方法研究及分析

Study and Analyze on Feature Selection in Text Categorization

对五种特征选择方法：文档频率、互信息、信息增益、期望交叉熵、统计作了简要的介绍，并且结合KNN分类算法，使用查全率、查准率、F1值对五种特征选择方法分别进行评估，提出并讨论了一种互信息修正的方法。

First this paper makes a brief introduction about DF, expected cross entropy ,MI, IG, andstatistic. Then combining with KNN classification algorithm, it assesses the four methods of feature selection by recall, precision, Fl. A% last, this paper proposes and discusses one method of improving MI.