Project Overview [Video]
In today’s fast-paced financial world, credit cards have become a staple for consumers. They offer convenience, enable instant purchasing power, and often come with rewards or benefits that make them attractive financial tools. With the rise in credit card utilization, financial institutions and credit card companies are in a constant endeavor to maintain profitability while ensuring a seamless experience for their customers.
A significant challenge these institutions face is the occurrence of “credit card defaults”. A default happens when a cardholder fails to make the required minimum payment on their credit card bill for more than a certain period. Such defaults lead to revenue losses and can impact the financial health and reputation of the lending institution.
The objective of this project is to predict potential credit card defaults before they occur. By anticipating defaults, the company can take proactive measures such as personalized communication, restructuring of credit limits, or even offering financial counseling, ensuring both the institution’s and the customer’s financial wellness.
Our aim is to develop a model that can accurately identify potential defaulters based on historical transactional and demographic data. This predictive insight will empower the institution with data-driven decision-making capabilities, enhancing customer relationships while ensuring sustained profitability.
We will perform EDA and Pre-processing on the Training Data, create a Random Forest Model, evaluate and optimise the Model using the Grid-Search Cross-validation technique and then also create a Decision Tree model and compare the results with the Randor Forest results.