Churn is when a customer stops doing business or ends a relationship with a company. To retain valuable customers and get ahead of the competition, it's common to build a customer churning prediction model to take proactive action.
Objective: create a customer churning model Source: Telco Churn Dataset
Outline: Exploratory data analysis Feature engineering/selection Normalization of features Modeling (SMOTE for imbalance data; as the churn and not churn data are imbalanced, SMOTE is used to oversampling the minority data) Evalualtion of performance Model tuning
Please see the following link if the results are not shown in Github: https://nbviewer.jupyter.org/github/fylinhub/Predicting-Customer-Churn/blob/master/predict_customer_churn.ipynb