🧠 Student Burnout Prediction using Machine Learning
This project predicts whether a student is at risk of mental burnout using behavioral data such as study hours, screen time, sleep patterns, and physical activity.
✨ Project Highlights
- 📊 Built a dataset of 30 students with 5 behavior-based features
- 🤖 Trained a Logistic Regression model using scikit-learn
- 🎯 Achieved 83.3% accuracy on the test set
- 🔥 Visualized feature correlations using a heatmap
- 📱 Entire project developed on Google Colab using a mobile device
🛠 Tools & Technologies
- 🐍 Python (Pandas, NumPy, scikit-learn, Seaborn, Matplotlib)
- ☁️ Google Colab
- 💻 GitHub
📊 Results Overview
| Metric | Score |
|---|---|
| Accuracy | 83.3% |
| Precision | 0.83 (Non-Burnout), 0.00 (Burnout) |
| Recall | 1.00 (Non-Burnout), 0.00 (Burnout) |
| F1-Score | 0.91 (Non-Burnout), 0.00 (Burnout) |
👤 About Me
Name: Kaviha R. M
Degree: B.E. CSE
College: V.S.B College of Engineering Technical Campus
Email: kaviharavichandran2006@gmail.com
GitHub: github.com/kaviha2006
🚀 This project was built as part of my learning journey in Machine Learning, and I’m excited to share it while applying for the Amazon ML Summer School 2025.