Shang Xinping and Wang Yi, Dongguan City University, China

With the rise of Internet finance, the competition of banking industry is becoming increasingly fierce. To gain more accurate and comprehensive insight into customer needs and improve customer loyalty, it is essential to establish a customer churn analysis model. This kind of model can help banks identify customers who are about to lose, facilitate business decisions, retain relevant users, and ensure that bank interests are not affected. Under this background, this paper establishes a customer churn prediction model using ensemble learning algorithm. Experimental data show that the model can effectively predict and analyse the loss of bank customers.

customer churn; data preprocessing; XGBoost.