Predicting-Customer-Churn-For-a-Telecom-Company

Business Objectives:

1. Identify customers likely to churn in order to guide retention strategy
2. Minimise false churn predictions to avoid unnecessary spend on loyal customers and thus optimize resources directed for retention.

Models Evaluated:

1. Logistic Regression (Unpenalized)
2. Ridge (Logistic regression with 'l2' penalty)
3. RidgeClassifier
4. Lasso (Logistic regression with 'l1' penalty)
5. Random Forest
6. Support Vector Machine
7. MLP Classifier
8. K_Nearest_Neighbor
9. Gradient Boosting Classifier