A study on bank satisfaction based on data mining technique

Abstract

Service satisfaction survey is important reference index for many enterprises in customer relationship management. The survey result, through all kinds of data mining techniques, can be used to understand the drawbacks of an enterprise on customer service and what are the major factors affecting these drawbacks, and the result can also be used as reference for improving the relationship between enterprise and customer in the future. First, this article will cite the survey result for service satisfaction performed by Louis Harris International Taiwan on 40 privately owned and government-owned banks in Taiwan area and perform grey relational analysis; meanwhile, the Grey Relational Grade of the analysis result will be ranked so as to understand the banks with satisfaction performance rankings in the top three rankings and last three rankings; then decision tree will be adopted to analyze what are the major factors affecting satisfaction performance, and it is hoped that the result can be used as reference by the bank managers; finally, finally in this article, three data mining techniques such as genetic evolution neural network, Back-Propagation Neural Network and decision tree will be used to set up bank’s service satisfaction performance detection model. From the analysis result, it can be seen that the top three rankings in satisfaction performance are respectively E.Sun Bank, Union Bank of Taiwan, AsiaPacific Bank, and the last three rankings are:Taiwan Cooperative Bank, Chiao Tung Bank, ChinFon Bank; the major factor affecting satisfaction performance for all the banks is service speed, for the privately owned bank, it is the service attitude of the bank service personnel; among three data mining techniques, genetic evolution neural network model has the best detection capability.