Abstract
With the rapid development of the technology of data warehouse and data mining, customer relationship management (CRM) becomes more and more important.
On this need, using the SEMMA frame based on Data Mining(DM) is introduced in this paper. At last the model which is found by TreeNet is chosen because of its good evaluation index. SAS and TreeNet are used to complete the design and realization of PHS churn model in Hangzhou. Evidence shows that basic monthly fee have significant influence.
Combined with theory of DM, the author tries to make some advice and recommendations for the decision maker in the PHS churn of main factors and customers of the detaining means etc. This predictive model is of significance in solving the problem of predictive telecom churn.