Beta-Bank
Beta Bank's customers are leaving: little by little, slipping away every month. The bankers have discovered that it's cheaper to save existing customers than to attract new ones. We need to predict whether a customer will leave the bank soon. You have data on past customer behavior and contract terminations with the bank. Build a model with the maximum possible F1 value. To pass the review, you need an F1 value of at least 0.59 for the test data set. Also, measure the AUC-ROC metric and compare it with the F1 value.
Features
-
RowNumber - data string index
-
CustomerId - unique customer identifier
-
Surname - last name
-
CreditScore
-
Geography - country of residence
-
Gender
-
Age
-
Tenure - length of service for the customer
-
Balance - account balance
-
NumOfProducts - number of banking products used by the customer
-
HasCrCard - client has a credit card (1 - yes; 0 - no)
-
IsActiveMember - active customer (1 - yes; 0 - no)
-
EstimatedSalary - estimated salary
Objective
- Exited - customer left (1 - yes; 0 - no)