Based On Data Mining Technology A Commercial Bank Customer Relationship Management

The core of customer relationship management in commercial banks is to use the advanced management methods and information technology, to integrate customer resources through the restructuring of business processes, then to realize the customer’s information and resource sharing at the use of a Client Management System, to provide customers with quality products and service, maintain and develop more customers, and ultimately achieve the maximization of corporate profits. In recent years, with the development and mature of data mining technology, data mining is more and more used in commercial banks’customer relationship management, to distinguish the existing customers so that they can explore the key customers of the banks and their potential requirements, to provide one-to-one personalized service, which greatly improves customers’ satisfaction and loyalty, not only attracts more and more new customers, but also avoids the loss of old customers. Moreover, the marketing of products and services becomes more targeted in commercial banks at the help of data mining, because the commercial banks only provide the products or services to potential customers, which effectively improve the success rate as well as efficiency of selling products and services, thus reduces marketing costs.In this paper, in order to predict whether the new customers of a bank will buy financial products, the paper uses decision tree model in data mining technology to identify the characteristics of customers who have purchased financial products (the amount of the deposit, age, gender, education, years of holding cards) to classify and predict new customers’demand for financial products with the use of this feature. First, this article discusses in detail the classification principle, the theory and algorithms of decision tree, the advantages of decision tree model, and the decision tree pruning classification rules. Then, the paper discusses the accuracy and scalability through two aspects—interpretation and verification. Finally, according to the results of the model analysis, the paper successfully uses the implementation of the classification of the target customers, to provide a scientific basis for the commercial banks’personalized service and relationship marketing.