Customer Churn Prediction based on Stacking Model

In the process of product marketing, companies are likely to face the serious problem of losing their stock of customers. However, analyzing churned customers through data mining techniques can not only retain those customers who may be lost but also help companies target product optimization and management improvement. In this paper, a bank credit card customer churn prediction model is built using the Stacking model to predict the churn of credit card customers. The model has higher accuracy than XGBoost, random forest, and logistic regression models, runs well, and has better customer management implications.