/Machine-Learning

Check my projects related to ML feature engineering and modeling.

Primary LanguageJupyter Notebook

Predictive Modeling for Customer Retention in the Banking Sector

Problem and General Objective

This dataset is originally found on the website https://leaps.analyttica.com/home, but it was obtained from https://www.kaggle.com/sakshigoyal7/credit-card-customers.

  • The manager of a bank branch is worried about a growing number of customers leaving their credit card services
  • The goal is to proactively predict which customers should be targeted for improved services to reverse their decision to leave
  • The dataset contains information on 10,000 customers, including age, salary, marital status, credit card limit, and category
  • Only 16.07% of customers who have left the service are included, making the dataset imbalanced
  • Transformations will be necessary to address the dataset's imbalance

Specific objectives:

  • Identify behaviors and characteristics of individuals who tend to leave or stay with the bank's service
  • Generate a model that can predict whether a customer will leave or stay with the offered service
  • Propose short and long-term actions that can better retain customers who are leaving based on the obtained information