Research Of Q Bank Customer Relationship Management System Based On Data Mining

The bank customer relationship management(CRM)system is a comprehensive system,which is a combination of computer hardware and software,customer relationship theory and bank network platform technology.Its purpose is to count,classify,and extract the customer information,and to form a model for the customer,which help the bank keep the original customers and find more potential customers,and provide related services and products according to the actual needs of the customer.Based on the operating situation,developing trend and existing problems of Q bank in Xi’an,Shaanxi,a CRM system based on data mining,database,customer management and software engineering is designed in this thesis.The research content of this thesis mainly includes the following aspects:(1)Firstly,in view of the research status of the CRM in domestic and foreign banks,especially for the problem of establishing the data platform for user relationship,we have made a detailed description and introduction for the management platform designation,adopting the technology,and data the existing problems etc.The description provides relevant data and technical support for the design of the CRM system of Q bank,the implementation of data processing algorithm and the usage of customer data.(2)Secondly,through the analysis of the transaction behavior and holding product of customer in Q bank,a new K-means clustering algorithm is proposed to classify the different customers.K-means clustering analysis algorithm based on traditional improvements include two aspects: to solve the local optimal solution,the algorithm selects the fixed analysis of the center of the initial K value,replacing the traditional algorithm of random K value,according to the characteristics of the bank customer,this thesis chooses K = 7;to solve the problem of noise sensitive,the algorithm selects the cluster center points after each of iteration,and defines the compactness degree using the absolute error standard instead of using square error standard.The proposed method can classify bank customers effectively,which provides strong support for banks to differentiate services and products according to the customer categories,and promote sustained and stable development of banks.(3)Lastly,a product sales method based on association rules is studied.According to the customer’s purchase behavior and consumption characteristics,it can find out the relationship between different products and the connection points between different customers from a large number of product transaction records,which help the bank to provide individual and diversified products for existing customers,reduce the cost of marketing and improve bank profits by cross-selling.To sum up,the Q bank CRM system based on data mining technology is designed,the designed CRM system is on the basis of the Q bank’ management characteristics and actual needs,which also combines with the basic principles of customer management.The effectiveness of the system is proved by the testing system.