A Research In Customer Attrition With Data Mining Technology

The customers are the focus of competition between the mobile telecommunication companies. How to efficiently control customers from changing net becomes urgent with saturated capability of the market, relatively stable price of the terminal products and telecom service and slowly increasing of new customers. Reducing the changing net rate and the negative growth has become a difficulty to figure out for the mobile telecommunication companies in no time.Based on the data mining theory, with SPSS tools utilized and stress laid on the service understanding, data extract, data cleaning, data conversion, model building and application, a detailed research into the consumptive behavior of sampled Nanjing mobile communication subscribers, an investigation of the consumptive feature of changing net users are given ,the forecast model of the probability of customers to change net is modeled, and a overall solution is proposed from the aspect of marketing and management to fulfill the forecast model of changing net, and improve the applicability of the model.After the forecast model of changing net created, the new data source is used to validate it. The results prove that the model has the better application performance. It can make the corporations organize some activity to save the customer of the high changing net rate specially, reduce the loss of changing net customers and benefit themselves. At the same time the process of the data mining is also useful for other telecommunication industry to analyze the customer losing.