MindMatric is an intelligent system that predicts stress levels in percentage using screen time data and social media activity.
It applies machine learning models to analyze behavioral patterns and provides personalized suggestions for stress management.
Collaborative Project
Original repository maintained by @RohitDev96
My contribution focuses on model building and backend development.
- 📱 Stress prediction using screen time & social media behavior
- 📊 Stress level output in percentage
- 💡 Personalized stress-reduction suggestions
- 🤖 Machine Learning–based prediction
- 🌐 Available as Web App and Android App
- ⚡ Fast and scalable FastAPI backend
- 📌 Sources:
- Kaggle datasets
- Real-time collected data (CSV format)
- 📊 Features used:
- Screen Time
- Activity Level
- Mood
- Emotional State
The system uses the following models:
- Random Forest Classifier
- Logistic Regression
These models analyze behavioral patterns to predict stress levels effectively.
- Python
- Java
- TensorFlow
- Pandas
- Scikit-learn
- Matplotlib
- FastAPI
- HTML
- Web Application
- Android Application
- 🎯 Stress Level: Percentage (0–100%)
- 📝 Suggestions:
- Reduce excessive screen time
- Improve daily activity balance
- Mood-based stress management tips
- ✔ Machine learning model building
- ✔ Model training and evaluation
- ✔ Backend API development using FastAPI
- ✔ Integration of ML models with frontend
- Developed as a collaborative project
- Main repository maintained by @RohitDev96
- Contributions clearly documented for transparency
✅ Completed and fully functional
📱 Deployed as both Android App and Web Application
- Deep learning–based stress prediction
- Real-time mobile sensor data integration
- Personalized long-term mental health analytics
- Cloud deployment and scalability
- Kaggle for dataset resources
- Open-source ML libraries
- Team collaboration and support