Bank Card Customer Classification Analysis System Design Based On Clustering Algorithm And Realization

Accurate custom classification is the foundation of an efficient customer relationship management of modern banks. The custom classification aims to divide the customers into different clusters, which helps the banks to analyze and predict the consume pattern of customers and consequently establish better services system for customers. It also helps the banks to perform different management for different customs and thus improve the competitiveness and profitability. Because of the potential factors that influence the customer classification, one should choose some schemes with better performance in classify the customers. The data mining technology can extract the implicit knowledge and spatial relationship patterns, and then divide the features of the customer data into several clusters.Based on the K-means algorithm, which is a simple and efficient clustering algorithm in the data mining technology, we design a classification system for the bank customer. The system can be divided into several subblocks including the preprocessing, data mining, data management and system maintenance blocks. The system can not only provide various kinds of direct and flexible presentation ways for the customer data, but also provide the data updating and management function for the administrator. After a designation of the data flowchart of the system, the realization of the subblocks and their relationship is analyzed, especially the data mining using the K-means algorithm. The system is realized under the environment of , and finally its efficiency in classification of customers is proved by simulations.