In this project, I use supervised learning models to identify customers who are likely to churn in the future. The objective of this project is to help banks identify customers who are at risk of leaving, so that they can take proactive measures to retain them. Furthermore, I analyze top factors that influence user retention.
● Predicted customer churn probability based on labeled data to increase customer retention.
● Preprocessed datasets by data quality checking (completeness, consistency, outliers, etc.), data cleaning (removed missing,
irrelevant, and duplicated data, etc.), feature selection(correlation matrix), and feature engineering (one-hot encoding for
categorical features and standardization for numerical features).
● Trained multiple supervised ML models (Logistic Regression, KNN, Random Forest, and K-Nearest Neighbors )
● Evaluated the models using confusion matrix, accuracy, ROC_AUC, and F1 score of classification via k-fold
● Increased the customer retention rate by offering promotions to the customers who were predicted to churn.