Customer Churn Prediction

Customer churn in the telecommunications sector presents significant challenges, impacting profitability and market share. This project leverages a hybrid ensemble modeling approach combining logistic regression and XG Boost classifiers to predict churn. Key drivers such as satisfaction scores, tenure, and billing details were identified using exploratory data analysis and feature engineering. The ensemble model demonstrated high predictive accuracy, aiding in targeted retention strategies and operational efficiency optimization. The research highlights the utility of machine learning in addressing customer churn and proposes enhancements for scalability and real-time application in telecom operations.