The Research On Data Mining Of Chinese Text

With the great development of Internet and the elevation of corporation's information degree, more and more information are being accumulated, most of which exist as a form of text. Thus, text mining comes into being as a new topic of data mining and a widely disputed research aspect, which arouses great interest. The research of Chinese text mining is at its early stage, till now there is no formal text-mining system for Chinese text. There exist many problems, and we study them in this paper.First, we discuss Chinese Text Mining (CTM) in theory. We present the definition of CTM based on the concept of Data Mining. We describe the disposal process of CTM by analyzing its characteristic, then, classify it from the aspect of function. In this way, CTM can been understood wholly.Second, by analyzing existing text classification technology, we discuss Chinese text classification at an angle of Text Mining, which includes word segmentation, characteristic extracting, characteristic matching in Chinese text, etc, and accomplish a system(STCS).Third, we utilize traditional association rules into text field, and present the definition of text association rules, query language and model expression. At the same time, we put forward a text-association rules algorithm (MATA) based on Aprior and IMAARC.Finally, we simply discuss two applications of Text Mining.