
Focus Management and Digital Wellness Platform
TypeScript
Tailwind CSS
Python
Firebase
Three.js
Framer Motion
Next.js
Overview
This is a detailed description of Project 1. It includes information about the project's goals, challenges, and solutions.
Features
- π§ Predictive Nudges β An XGBoost model predicts distraction risk in real-time with over 97% accuracy and provides proactive nudges.
- π¬ AI Productivity Coach β Integrated with Google Gemini to give actionable, personalized tips for staying on track.
- π Focus Tools β Pomodoro timer + daily/monthly/annual goals to help users plan and track productivity sessions.
- π§ Brain Visualizer β Live hemisphere visualization showing the balance between logical and creative activity.
Technical Details
- Frontend: HTML, CSS, JavaScript
- Backend: Python Flask (for distraction risk prediction)
- Machine Learning:
- Model: XGBoost
- Data: 50,000+ labeled entries from user interaction sessions
- Training: Done in Google Colab
- Split: 80% Training, 10% Validation, 10% Testing
- Model Files: distraction_model.pkl, feature_columns.pkl
- Deployment: Firebase (for hosting demo website)
- AI Chat Assistant: Google Gemini API