Credit Card Default Prediction
Overview
This is an end-to-end data science project to predict the likelihood of a credit card client defaulting. The dataset used can be found here.
Key Techniques : Data Science Foundations, Operations Research, Statistical Modeling
Links :
Main Challenges
- Understanding the data science pipeline from data cleaning, preprocessing to modeling using statistical knowledge and machine learning algorithm learned in BT1304 Optimization Methods for Business Analytics.
My Contributions
- Outlined the project pipeline and milestones.
- Wrote 90% of the code for the notebook. Code is well-documented and decisions of model selection and evaluation are explained clearly.
- Worked on the project report with two other members.
Tech Stack
Python libraries - Scikit-learn, Matplotlib, Numpy
Outcome & Impact
This project is awarded full marks (20/20) for BT3104 module in Y2S2. Instructor’s feedback for each section can be found here.