Feb 12 2023 | Hugo Hiraoka | hhiraokawatts@gmail.com
Classification models to predict credt card customer attrition using Logistic Regression, Decision Trees, Random Forest, Bagging, Gradient Boosting, Adaboost, and Xgboost, and performs hyperparameter tuning using Randomized Search. This is a supervised learning model.
First National Bank has seen a steep decline in the number of users of its credit cards. Like other banks, credit cards are also a good source of income for Thera Bank because they generate revenue by collecting annual fees, balance transfer fees, cash advance fees, late payment fees, foreign transaction fees, and others.
When a credit card customer leaves, the bank will lose that revenue source, so it is very important that the bank implements measures to avoid losing those customers.
To achieve this, we will implement a classification model that will help First National Bank improve its services so credit card customers do not leave the service.
Finally, we will generate a set of insights and recommendations that will help the bank reduce credit card customer churn.
- What kind/type of Credit Card customers churn?
- What things in common do these Credit Card customers have?
- Why do these Credit Card customers churn?
- What can be improved to decrease Credit Card customer churn?