CUSTOMER CHURN PREDICTION IN BANKING INDUSTRY

Abstract

ABSTRACT In the present era of high competition, all businesses have realized that retaining their existing customers is the most important strategy to survive. I has been shown that it is less expensive to retain and satisfy the exist customers than to acquire new customers. In this context, predicting and managing churn has become critical to the survival of the company. This paper uses Support Vector Machines (SVMs) in order to predict churn in banking industry. The analysis was carried out using customer data of current accounts of a large public sector bank in India. With very limited access to the demographic details of the customers, the paper used behavioral variables extracted from the pattern of credit and debit in the account for a period of 12 months. It was found that the churn can be predicted with a high degree of accuracy using these behavioral variables. Keywords Customer Churn, Banking Industry, Support Vector Machines, Data Mining Techniques