Application Research Of Data Mining On CRM System Of The Tobacco Business

After joining WTO, the monopoly system of the Chinese tobacco will undergo drastic changes. The gradual involvement of foreign tobacco groups in the market may cause fierce competition with our own tobacco enterprises. They will compete in many fields, such as brand, price, talent, from which a more important one is the sense of serving. The competition of services is the competition of customers. The one who gains the customer's satisfaction will gain more advantages in the market competition. It has become an urgent issue that how to apply advanced computer technology to the tobacco client relation management (CRM), make the most of the customer's potential, and offer the tailored one-to-one service to customers.This dissertation uses data mining technology to deal with tobacco business customer relationship management. Detailed work is as follows:(1) Firstly, the concepts of different types of CRM have been introduced. Then, the necessity of applying CRM Systems to tobacco enterprises has been discussed. The business and the functional model of the tobacco system have been analyzed and the supporting technology of tobacco enterprise CRM system has been summarized.(2) Finally, the order data and customer information data have been taken as data source. The customer group has been subdivided by using the method of the cluster. Data is pre-processed to get the satisfied data for data mining .The RFM analyzing model has been established firstly, then the k-means clustering algorithm has been adopted to conduct the clustering analysis for the three fields—"Regency", "Frequency" , "Monetary" , and the group of subdivided customers can be gotten finally. Experiments exhibits good performance of the whole system.