Segmentation of the Bank Client Value Based on Fuzzy Data Mining
First, this paper will analyze the connotation of individual bank clients’ value in-depth. Based on this foundation, the main factors that reflect the client value will be selected. Then it will use the expert method to build a client value evaluation sys
- PDF / 155,511 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 32 Downloads / 144 Views
Segmentation of the Bank Client Value Based on Fuzzy Data Mining Hong-bo Wu and Ming-hui Guan
Abstract First, this paper will analyze the connotation of individual bank clients’ value in-depth. Based on this foundation, the main factors that reflect the client value will be selected. Then it will use the expert method to build a client value evaluation system. Finally, using the fuzzy data mining algorithms, this paper will obtain the basic model of bank client value segmentation, which will provide the basis for effective prediction of high-value clients. Keywords Fuzzy data mining • Customer value • Evaluation system
Introduction With the fierce competition in the financial industry, how to effectively analyze the client data has become the key of the bank-client relationship management (Adrian and Sue 2001; Wolf 2001). Analysis showed that the cost of attracting new clients is five times of retaining old clients. Therefore, tapping high-value clients from existing clients will become the central force driving the development of banks in the future. Since the relevant data of bank clients possess a high degree of complexity and ambiguity, it is difficult to use the conventional method to accurately predict the high value. Aiming at this issue, this article advises to breakdown the bank clients into segments from the client value point of view (Frederick 1996), and then predict the high-value clients. The study found that, at current stage, there are two ways to study the client values: One is the client segmentation based on the life cycle value model (Hwang et al. 2008). However, this method requires the estimation and calculation of all the costs and income related to the clients within the bank
H.-b. Wu • M.-h. Guan (*) Department of Information Management and Information System, Harbin University of Science and Technology, Harbin, China e-mail: [email protected] R. Dou (ed.), Proceedings of 2012 3rd International Asia Conference on Industrial Engineering and Management Innovation(IEMI2012), DOI 10.1007/978-3-642-33012-4_56, # Springer-Verlag Berlin Heidelberg 2013
569
570
H.-b. Wu and M.-h. Guan
throughout the life cycle, which definitely increased the difficulty of evaluation (Boyce 2000). The other way is to create a client value assessment system (Qi Jiayin et al. 2002). Although this method can either analyze the clients’ current situation or emphasize the importance of clients’ potential of development on the banks (Quan Mingfu et al. 2004), the analysis of combining the AHP method (Wang Fengliu et al. 2009), the model method (Qian Weining and Zhou Aoying 2002), and the K-Means evaluation methods (Deng Guangli et al. 2005) showed that some only stayed in analyzing certain aspects of the client value, and some segmented the client but only getting a rather vague result. None of them could predict the high-value clients generally and comprehensively. As a result, this paper first analyzed the main factors of client value, and then designed the evaluation indexes for the bank client value. Fin
Data Loading...
