Bank Customer Churn Prediction - Machine Learning

A machine learning model that predicts which bank customers are likely to churn, enabling targeted retention before customers leave. The project covers the full workflow: exploratory data analysis, preprocessing, model training and comparison, and evaluation.

Multiple classifiers were trained and compared, with the best model achieving 85.7% accuracy. Beyond raw accuracy, the analysis examines the features that most drive churn , turning the model from a black box into actionable insight about why customers leave.

Built with: Python, pandas, scikit-learn

Applicable to any subscription or customer-retention setting where identifying at-risk customers early has clear business value.