Recommendation Of Discount Package To Churning Customers Based On Data Mining

Along with the rapid development of communication industry, the competition among the carriers is getting more and more drastic. Customer resource has become the key of enterprises' competition. Nowadays almost each carrier faces a serious problem—customer churn. Just for a great quantity of customers' vanishment, it has brought great loss to many companies. And as the communication market becomes saturated, acquiring the new customer is getting much more expensive than retaining the existing customer. In such a severe circumstance, it has become one of the focuses of operation corporations that how to avoid customer churn and carry out retainxnent.In order to face the competition , the service provider of communication industry, are transferring from "Service-driven" business to "Market-driven" one, in other words from "Technology-driven" business mode to "Customer-driven" one. The carriers provide the services and products in order to satisfy personal demand of customers. Discount package take a large part among all these services and products. Carrier provide tens of different discount packages for customer to choose. They can economize money for customers in order to sustain their loyalty. But sometimes the type of discount package that customer choosed is not the cheapest.Then in this thesis , according to the actual practice of discount package of national carriers, we get the best one for customer churning to follow the changing of the discount package. Following are the main jobs:1. Learn the relevant knowledge of communication industry , discount package and data mining .Load data into oracle DBMS and pre-process the data.2. Analyze the behavior of customer, process operational data according to discount package model and get the best one.3. To explain and evaluate the result, verify the the operational and validity of the model.