Showcase of example on making credit scoring prediction from various informations among several tables in a database (Classification problem).
Tools and Techniques used: Pandas data preprocessing, PPS (Predictive Power Score), PCA (for Dimensionality Reduction), Halved GridSearch CV (for model tuning)
Models used: Logistic Regression and XGBoost
Initially, this was used for technical test in BNI (as a Data Science ODP in 2021) and Home Credit Indonesia (as a Junior Data Scientist under Data Science Academy in 2021).
This personal project was forked from https://www.kaggle.com/c/home-credit-default-risk