文本分类中特征提取的代码。采用信息增益法，对文本的空间向量模型能达到有效降维。文件的输入形式必须是词号-词频形式。-

Text Categorization feature extraction code. Using information gain method, the vector space model of the text to achieve effective dimensionality reduction. Enter the file must be in the form of word number- word frequency form.