| Along with the world economics globalization, the market internationalization and the national reform, the domestic telecommunication market environment is gradually reasonable and the competition is day by day intensifier. The customer average lifecycle reduction has seriously afected the telecommunication enterprise's development.In this case, how to maintain current customers and predict chum to adopt suitable marketing measures and to prevent loss become an urgent problem for telecom operators. The focus of data mining technology has turned to various application domains after several years' development. Telecom is typical industry with dense data which hides a great deal of valuable information for enterprise.The aim of the thesis is to get data mining model based on historical customer data which can generate the customer list with high probability to leave the company. It makes use of the WHS's history data to establish a customer chuming model. It introduces CRISP-DM standard data mining process model. Combining the characteristic of business data, it emphasizes on the data mining methodology of customer chum in telecom industry, puts forward the specific and efective resolving solutions for the problems involved in business understanding, data understanding, data preparation, modeling, evaluation and deployment. The main comtents include:1. Analyse and study the decision tree's arithmetic and describes in detail the principle and process of Microsoft decision trees algorithm.2. Establish WHS's customer chuming model: discovers the characteristic of churned customers using the data mining technology, builds the forecast model of customer chum and applies the model to forecast the churning customers. 3. Realize WHS's customer chuming forecast system: complete the general design and general module of the system, realize the data management module and data stream.With the progress of data mining technology, more and morevaluable customer's information will be found and will direct the telecommunication company to make business decision. |