Machine Learning Based Churn Prediction

Computer Engineering

Customer turnover is a significant issue and one of the most pressing challenges for big businesses. Companies are working to create methods to forecast prospective customer churn since it directly impacts their revenues, particularly in the telecom industry. As a result, identifying factors contributing to customer turnover is critical to taking the required steps to decrease churn. Our work's essential contribution is developing a churn prediction model that helps telecom carriers estimate which customers are most likely to churn. The model created in this paper employs machine learning methods on a large data platform to provide a novel approach to feature engineering and selection. This research also established churn characteristics that are critical in discovering the fundamental causes of churn to gauge the model's performance. CRM may enhance productivity, offer suitable promotions to a group of potential churn customers based on similar behaviour patterns, and vastly improve the company's marketing efforts by identifying the main churn drivers from customer data. The accuracy, precision, recall, f-measure, and receiving operating characteristics (ROC) area of the suggested churn prediction model are examined. Furthermore, using the rules created by the attribute-selected classifier algorithm gives causes behind the churning of churn clients.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.