With a great scale popularization of Internet and improvement of the information of corporation, therefore how to deal with these big capacities of online documents has been the important research subject. Technique of Text Mining can quickly and effectively abstract the useful information from the many data. Text Classification is one of the most important and popular techniques in Text Mining.Some basic concepts, relative underground of Text Mining,all the theories and each kind of methods are introduced firstly in this paper.The key techniques of Text Mining related including text describing model,feature abstracting, appraisal method and classification method are taken intensive theory and arithmetic description,and made a detail introduction of feature abstracting and Naive Bayes classification.In addition,This paper offer and take affect a kind Text Classification system based on improvement mutual information and supporting Naive Bayes,and takes affect the main process and technique problem in the system process.In end,doing some experiments of Text Classification system based on improvement mutual information and supporting Naive Bayes, experiment results prove the arithmatics and system being high classification ratio. |