Overview
This project was developed as part of the African Credit Scoring Challenge, a global competition with over 1,000 teams participating. Our team ranked 156th with a public score of 0.700, placing us in the top 16% of participants.
We built a mix of Machine and deep learning model to predict credit risk using various machine learning techniques, feature engineering, and model optimization.
Dataset
The dataset includes customer financial information, transaction history, and behavioral data to predict the likelihood of loan default. Due to competition rules, the dataset is not included in this repository.
Technologies Used
Programming Language: Python
Libraries: TensorFlow, Scikit-learn, Pandas, NumPy, Matplotlib, Seaborn
Tools: Jupyter Notebook, Google Colab