The Application Of Data Mining In Bank Customer Relationship Managemennt

With the deepening of financial reform, financial institutions continue to increase. In recent years, the Internet in the application and innovation of the financial industry is more and more fast, the popularity of smart devices shifting makes the whole social environment. Banking is inherently competition is intense, the development of the new situation and no small impact to the banking industry. Banks in order to maintain their own development, we must explore the core competitive advantage. While the customer is bank based on the fundamental, to update the bank customer relationship management is imperative. This article is precisely based on this situation, to explore in the popularization of the network circumstances, the bank how to use data mining techniques for customer relationship management, to build the core competitiveness.In this paper, the theory of customer relationship management related to the research, puts forward some new view of customer value, customer is the most important social network information into the whole process of customer relationship management (CRM), secondly discusses some of the data mining technology in the customer classification optimization method. Customer value is the basis and the core of customer relationship management (CRM). Previous customer value mainly through the identity of the customer and customer assets, customer value has been referred to in the future. This paper suggests, earlier for customer management, and the potential customers as an important measure, will be part of the social network information into the index of potential customers. In after the construction of the system of customer value, customer classification becomes the key step of the application of customer relationship management (CRM). In this paper, through the data mining technology in the clustering and decision tree method combined to realize the customer segmentation, based on the data mining process, namely before clustering using charting tools for exploratory data analysis, in order to better choose clustering center and categories. And put forward some optimization on the algorithm. Through bank telephone return visit to database and the customer data, and data processing, model optimization. Eventually making clustering results than the original model has been a certain degree of ascension.