Welcome to my Data Analytics repository — a curated collection of real-world projects and research experiences in predictive modeling, machine learning, and data-driven decision-making. These projects reflect my commitment to solving impactful problems through data science and analytics.
Duration: Sept – Dec 2023
Objective: To predict 30-day hospital readmission risks for diabetic patients using electronic health records (EHR).
- Conducted predictive analytics on 100K+ diabetic patient records.
- Built models to forecast patient readmission within 30 days.
- Identified key risk factors such as insulin level, number of diagnoses, and patient discharge disposition.
- Proposed targeted interventions to reduce readmission rates and healthcare costs.
- Python, Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn
- Logistic Regression, Decision Trees, Random Forest
- Data Preprocessing, Feature Engineering, Hyperparameter Tuning
- Evaluation Metrics: ROC AUC, Confusion Matrix, Precision-Recall Curve
- Achieved high accuracy and recall in identifying at-risk patients.
- Developed visualizations to support clinical decision-making.
- Suggested strategies for prioritizing patient follow-ups and improving hospital resource allocation.