My bank loan eligibility project analyzed 614 applications using Python and PostgreSQL, with data cleaning, feature engineering, and classification models. The Logistic Regression and Random Forest models achieved 85.3% accuracy, and Credit History was the strongest predictor.
Across telecom, grocery, and retail projects, I used SQL, Python, and Power BI to analyze customer churn, sales, and profit trends. I also built dashboards and applied forecasting methods to examine monthly sales seasonality and future demand.
