Developed a Random Forest Classifier to detect “Risky” vs “Good” taxpayers based on income, education, marital status, and urban background. Performed EDA, feature encoding, and evaluation with high accuracy. Showcases robust ensemble learning using Python, Pandas, and Scikit-learn.