Comparative Analysis of Machine Learning Algorithms for Customer Churn Prediction in Banking

Comparative Analysis of Machine Learning Algorithms for Customer Churn Prediction in Banking

Abstract

Acquiring new subscribers for telecommunications providers is significantly more expensive than keeping existing ones. However, the annual churn rate for these providers lies between 15% and 35%. This paper surveys how machine learning has been used to address this problem over the last five years. It examines models ranging from simple logistic regression to gradient-boosting ensembles and sequence-aware deep-learning models. Results are compared using the IBM Telco, SyriaTel, and UCI KDD Orange benchmarks and accuracy, AUC-ROC, and F1-score metrics. The paper also examines how these models handle class imbalance using SMOTE. Its survey of more than 30 papers indicates that gradient-boosting models such as XGBoost, LightGBM, and CatBoost dominate this domain. Explainable AI methods such as SHAP and LIME are also discussed as performance enhancers and regulatory enforcers.

Authors

Raj Shekhar Singh, Rahul Raj, Karan Raj, Garima, Ankesh Jha

Institution

Noida Institute of Engineering & Technology (MCA Institute), Greater Noida, India

PDF Link

View Full Paper