A Gradient Boosting classification model for identifying 30-day patient readmission risk using cost and clinical feature interactions.
The project focuses on model interpretability through feature importance and ROC analysis, supporting applications in healthcare analytics and early risk detection.