In this work the purpose of the model our plan is to build a model that predicts an individual customer will churn form the bank (or not) given various data points and information form a historical data.
The goal of this model is to help decision making framework for predicting if a customer will leave or not.
Banks will be able to predict the risk of the customer on whether they will leave or not.
We will take the data from Kaggle , and we plan to usse these classes and features:
| class | Description |
|---|---|
| Exited | Customer will leave |
| Not Exited | Customer will stay |
| Field Name | Description |
|---|---|
| RowNumber | Corresponds to the record (row) number |
| CustomerId | Customers ID's |
| Surname | Surname of a customer |
| CreditScore | Number between 300–850 that depicts a consumer's credit worthiness |
| Geography | Customer’s location |
| Gender | Male/Female |
| Age | Customer Age |
| Tenure | Refers to the number of years that the customer has been a client of the bank |
| Balance | Customer's account balance |
| NumOfProducts | Refers to the number of products that a customer has purchased through the bank. |
| HasCrCard | Denotes whether or not a customer has a credit card. |
| IsActiveMember | If the customer is active or not. |
| EstimatedSalary | Customer's salary. |
- Technologies: Python, Jupyter notebook.
- Libraries: Seaborn, Matplotlib, Pandas profiling, NumPy, Pandas, Sklearn.