This project is a machine learning-based web application that predicts the probability of diabetes using health indicators data.
The model is trained on the BRFSS 2015 dataset and uses a Random Forest Classifier to analyze key features such as BMI, blood pressure, lifestyle habits, and general health metrics.
The system includes:
- Data preprocessing and cleaning
- Handling class imbalance using oversampling
- Model training and evaluation
- Interactive user interface built with Gradio
Users can input their health data and receive:
- A diabetes prediction (Diabetic / Non-Diabetic)
- Probability score with adjustable threshold
This project demonstrates the integration of data science and web deployment for real-world healthcare applications.
