Customer churn prediction using machine-learning techniques in the case of commercial bank of Ethiopia

Abstract

The number of service providers is increasing rapidly in every business. These days, there is plenty of options for customers in the banking sector when choosing where to put their money. As a result, customer churn and engagement have become one of the top issues for most of the banks. In this paper, a method to predict customer churn in a Bank using machine learning techniques, which is a branch of artificial intelligence, is proposed. The research promotes the exploration of the likelihood of churn by analyzing customer behavior. random forest (RF), logistic regression (LR), gradient boosting classifier (GBC), extreme gradient boosting classifier (EGBC), and light gradient boosting machine classifier (LGBCMC) are used in this study. Also, some feature selection methods have also been done to find the more relevant features and verify system performance. The experimentation was conducted on the churn modeling dataset from Kaggle. The results are compared to find an appropriate model with higher precision and predictability. As a result, using the Random Forest model after oversampling is better than other models in terms of accuracy. The experimental result shows that the Light Gradient Boosting Machine classifier outperformed with an accuracy of 98%, a precision of 97%, and a recall of 100%, with an AUC of 99% than other proposed supervised machine learning algorithms with balanced datasets across all evaluation metrics.