Project Overview
This project uses a healthcare dataset to predict diabetes outcomes based on patient health indicators. The data was cleaned, normalized, and trained using Logistic Regression and Random Forest models to ensure accuracy and generalization.
Tools & Techniques
Python Pandas Scikit-learn Matplotlib Seaborn
Key Outcomes
- Achieved 87% prediction accuracy using Logistic Regression.
- Enhanced recall using balanced dataset preprocessing.
- Visualized results with ROC curve and confusion matrix.