本文通过对Bayes、KNN、SVM 应用于中文文本分类进行比较实验研究。



应用ICTCLAS 对中文文档进行分词，在大维数，多数据情况下应用TFIDF 进行



特征选择，并同时利用它实现了对特征项进行加权处理，使文本库中的每个文本



具有统一的、可处理的结构模型。然后通过三类分类算法实现了对权值数据进行



训练和分类。-Based on the Bayes, KNN, SVM applied to compare the Chinese text categorization Study. Application ICTCLAS word segmentation of Chinese document, in large dimension, multiple-data cases where application TFIDF feature selection, and also use it to realize the characteristics of weighted items, so that each text text library with a unified, capable of handling structural model. Three types of classification algorithm then weights the data for training and classification.