This repository showcases Data Science projects that tackle real-world problems through data cleaning, feature engineering, and machine learning. Each project includes a complete workflow from data preprocessing to model deployment insights, offering a practical demonstration of how data-driven decisions can transform businesses.
Goal: Categorize customers by Recency, Frequency, and Monetary (RFM) metrics. Outcome: Pinpointed High-, Mid-, and Low-Value customers, enabling targeted marketing and boosted retention. Link Here
Goal: Predict which customers might leave the bank. Outcome: Achieved up to 90% accuracy (Random Forest), revealing key drivers like Age and Balance, and guiding targeted retention. Link Here
Goal: Aid early detection of CKD through data science. Outcome: 98% model accuracy with XGBoost, underscoring significant potential for healthcare interventions and resource allocation. Link Here
Goal: Forecast which prospects will purchase insurance. Outcome: 93.5% accuracy using Random Forest, offering valuable insights for optimizing sales and marketing strategies. Link Here
Contact LinkedIn:
Thank you for exploring these projects! Each one illustrates a unique approach to solving industry challenges using data-driven methods, highlighting the power of analytics to shape better decisions and outcomes.