Analyzing customer churn in banking: A data mining framework

Abstract

Customer churn, the loss of customers to a business, is a significant challenge in the banking industry. Retaining existing customers is crucial for banks to maintain profitability and sustain growth. This paper focuses on analyzing customer churn in the banking sector. The study utilizes data mining and predictive analytics techniques to analyse customer behaviour, identify churn patterns, and develop predictive models. This research uses a data mining technique called Gaussian mixture model clustering-based adaptive support vector machine (GMM-ASVM) to forecast customer loss in the banking industry. By analyzing consumer competency and loyalty to the banking industry using GMM, this study predicts customer behaviour using a clustering approach. An accuracy of 98% was attained while classifying the clustering results using ASVM. This study gives bank administrators the ability to analyse the behaviour of their clients, which may trigger appropriate tactics based on engaging quality and increase appropriate actions of administrator capacities in interactions with customers.