Psbc Card Customer Segmentation Research Based On Data Mining

With the gradual profit reducing of traditional businesses, commercial banks pay more and more attention to develop intermediary business. As an important part of traditional businesses, credit card business has been the new focus. Owing to intense competence, our credit card business model has become based on customers, data, and information. This model’s premise and foundation is customer segmentation. Nowadays, domestic banks distinguish credit card customers mostly basing on experience or statistics methods. That is behind the development of business. With the depth use of data mining, using data mining methods to distinguish and manage credit card customers is more and more important.The subject of this paper is preliminarily study a credit card customer segmentation model of WenZhou PSBC. It uses the data mining to deal with the huge number of card data, then chooses a right method to help the bank design card products and services. At last, it build a customer based marketing strategy, improve customer satisfaction and value.In this paper, firstly, I simply describe the present situation and question in card business of Chinese banks, and put forward the subject of this paper is research on PSBC card customer segmentation. Secondly, it introduces the theory of customer segmentation, and situation of the commercial banks’customer segmentation. Then, I put forward that many domestic banks are still use the traditional segmentation method which is defective. Thirdly, it introduces the data mining technology, and common customer segmentation methods based on data mining. Forthly, it specific elaborated that the author uses data mining tool "Enterprise Miner" to analysis the data of WenZhou PSBC card customer by classification method. And then evaluate the results. Finally, it discusses the contributions, limitations and prospects of the subject.This paper uses the CHAID decision tree method for customer classification, and it builds a classification model. Then it tries to judge categories of new customers with their basic information. All the work is just tries to find a way to help managers of PSBC to make business decision. This paper has its value to improve competitiveness of PSBC.