Credit Card Default Prediction
A machine learning project to predict credit card default risk in Taiwan, focusing on probabilistic modeling and financial risk assessment.
During my AI internship at Edunet Foundation, I worked on a data-driven project focused on predicting customer default payments in Taiwan to enhance financial risk management. Leveraging a real-world dataset, I explored various machine learning algorithms to assess the probability of default rather than simply classifying clients as credible or not. This probabilistic approach offered more meaningful insights from a risk management perspective.
Through rigorous evaluation using metrics such as the K-S (Kolmogorov-Smirnov) chart, I identified XGBoost as the most effective model for predicting defaults. This project strengthened my skills in AI, data analytics, and model evaluation while also deepening my understanding of how machine learning can be applied to financial decision-making.
Project Gallery
