A machine learning web application that predicts the likelihood of diabetes based on diagnostic measures (Glucose, Insulin, BMI, Age). This tool is designed to provide quick, accessible health insights using a Support Vector Machine (SVM) model.
👉 Click here to view the Live App
Please Read Carefully: This project is for educational and informational purposes only.
- The predictions generated by this model are based on historical data and statistical algorithms.
This application was developed as part of the CSCE 5214 coursework. It demonstrates the end-to-end process of deploying a machine learning model, from training to a web interface.
- Real-time Prediction: Instant analysis of health metrics.
- Visual Feedback: Clear, color-coded results (Low Risk vs. High Risk).
- AI Integration: Provides lifestyle and diet suggestions using OpenAI (simulated or live).
- Interactive UI: Built with Streamlit for a responsive user experience.
- Input: The user enters standard health metrics (Glucose, Insulin, BMI, Age).
- Processing: The app scales these inputs to match the training data range.
- Prediction: A pre-trained Support Vector Classifier (SVC) analyzes the data.
- Output: The app returns a risk assessment and a probability score.
This project is a modified fork of an existing machine learning repository.
- Original Repository: https://github.com/Aditya-Mankar/Diabetes-Prediction
- Modifications: * Migrated frontend from Flask/HTML to Streamlit for better deployment.
- Added visual styling and interactive tabs.
- Integrated OpenAI API logic for personalized health tips.
- Improved error handling and state management.
Distributed under the MIT License. See LICENSE for more information.
LICENSE