Customer Churn Prediction with Machine Learning Approaches

Abstract

Customer churn prediction is a critical task for many industries, such as telecommunications, banking, and e-commerce. This paper presents a comprehensive survey of customer churn prediction methods, which are typically classified into three categories: statistical methods, machine learning-based methods, and deep learning-based methods. The survey focuses on each category, introducing the most relevant approaches of churn prediction, as well as their respective strengths and weaknesses. We also discuss the challenges and open research issues related to this field. Finally, we outline the future research trends in customer churn prediction in order to inspire new research ideas. RandomForestClassifier achieved the highest accuracy of 84.00%, outperforming other machine learning and Deep learning algorithms.