Mining the Retail Banking Customers Characteristics Using Data Mining Techniques

Subject Areas : Jounal of Marketing Management

J. Nazemi ¹ ()
P. Jafari ² ()
H. Hashemi ³ ()

Received: 2013-08-25 Accepted : 2013-08-25 Published : 2012-03-20

Keywords:

Abstract :

Deregulation within the banking industry's and unprecedented growth competition in new technologies, every day the importance of keeping current customers and attract new customers are added. This study presents a two-step model to identify characteristics of different groups of bank customers, based on their profitability. The new criteria introduced to analyze the profitability of each customer. And then, this criterion has been used for clustering customers based on their profitability. Because in k-mean algorithm there is not a general rule for the optimal number of clusters and the number of clusters depend on the problem, therefore firstly with applied by Two-step clustering algorithm determine the optimal number of clusters. In this study, Customers clustering to 3 groups as golden, silver, and bronze customers. Then, by using K-mean algorithm different groups of customers are identified. And with help of Apriori algorithm association rules of each cluster is inference. The result of this exploration, help to banks have better understanding of current and future customer expectations. And through this may be facilitated develop of marketing strategies to attract and retain profitable customers.