Operators are facing more and more serious compliants such as telecom. In particularly, upgrade compalints is growing with each passing day. With improving competitiveness, solving the serious complaints has no time to delay. This article will make text mining with data of telecom by using Naive Bayes classification algorithm to identify upgrade complaints,forecasting the possibility of upgrade complaints. By useing R, one of the tools of data mining, according to the characteristics of the data,the appropriate algorithm be selected to analysis. The first two months data for the training set, the last two months data for the testing set. Paper will eventually be starting from the effect of text processing and classification test to the industry put forward the corresponding suggestion, analysising and give evaluation on research methods of this article. |