Enhancing Customer Churn Prediction in Digital Banking using Ensemble Modeling

Abstract

There are many purchases in the digital banking platform occurred daily. The electronic banking contains various transactions with different purchase behavior. Customer attrition from a digital store to another has become a challenge to the business owners. So, Businesses should measure their customer churn rate at regular intervals, as it is an important metric. Digital Banks started in building intelligent models to increase the customers satisfaction. Customer churn prediction (CCP) is an important strategy involved in the customer relationship management (CRM) strategies to forecast the probability of their attrition. This paper presents a classifier-based model to predict customer churn behavior based on the customer profile data. In this research, several supervised classification algorithms have been applied and compared like, KNN (k-Nearest Neighbors), Logistic Regression, AdaBoost, Gradient_Boosting and Random Forest. A unified voting is next applied for the set of classifiers in order to produce a high prediction accuracy. The model performance is enhanced by applying hyperparameters optimization and tuning. The most important features are defined and ranked using the Random Forest model. In the presented experimental case study with a dataset of size 10K, the model achieved an accuracy value 87%.