Customer Churn Prediction Using Machine Learning

Developed a machine learning model to predict customer churn using Random Forest and XGBoost algorithms. The project involved data preprocessing, feature engineering, and answering key business questions to provide actionable insights. Achieved an accuracy of 96% on the test data, validated through cross-validation. The model’s performance was evaluated using a confusion matrix, showcasing a balanced prediction of churned and non-churned customers.

Key Outcomes:

• Accuracy: 96% accuracy on the test data
• Business Insights: Provided answers to key business questions such as identifying cities with the highest churn rates, effective offers to reduce churn, and key churn reasons
• Modeling: Used cross-validation, and confusion matrix for performance evaluation