基于TFIDF的文本特征选择方法

A Text Feature Selection Method based on TFIDF

本文在分析比较几种用于文本分类的特征选择方法的基础上,提出了一种基于术语频率和逆文档频率的特征选择方法TDF。采用KNN和NaiveBayes两种分类算法对该方法进行了测试。实验结果表明,TDF方法较其他几种方法有较好的分类精度。

This paper compares several feature selection methods in text categorization, proposes a new feature selection method based on term frequency and inverse document frequency. We evaluate the effect of feature selection by using KNN and Naive Bayes classifters. Experiments show that the method can gain better effect.