Research On Bank Satisfaction Based On Data Mining

With the coming of information explosion era, people are paying more attention to how to obtain useful knowledge in the data, thus the birth of an emerging interdisciplinary field of data mining. Data mining algorithm to optimize and develop, making the application of data mining is more and more widely, such as the study of bank customer satisfaction. Nowadays increasingly fierce competition in the banking sector, only to improve customer satisfaction will make the bank stable long-term growth in the market.The content of this paper is to study the bank customer satisfaction with data mining technology. Firstly,the article introduces the definition and sources of data mining, and several common methods of data mining. Secondly, this paper introduces the bank customer data. Raw data is obtained through questionnaire survey. Analyze customer’s data variables, the data variables for segmentation to get the breadth and depth of the subdivision variables.7 main factors are obtained by using factor analysis method to reduce the dimension of 29 depth subdivision variables. Finally, a respectively decision tree model is established for the depth and breadth subdivision variables. The classification accuracy of the two models are relatively high and close. From the results of the breadth of the decision tree model can be aware that even the same bank, different branches and outlets have different customer satisfaction, as well as the gender has no effect on satisfaction evaluation. The depth segmentation model shows that the service specification has the greatest impact on customer evaluation, followed by lobby manager service, self-service equipment service, however, the teller service in the last.Through the data mining technology, analysis of customer satisfaction. It is convenient for banks to know the characteristics of the clients. To meet customer’s demand, corresponding service to different customer groups. At the same time, according to the results can make the corresponding marketing plan.