PhoneWatch AI — Teen Phone Addiction Detection

A production-grade Flask web application with ML-powered teen phone addiction prediction.

Project Structure

teen_addiction_app/
├── app.py                          # Flask backend
├── knn_model.pkl                   # Trained KNN model
├── processed_data.csv              # Processed dataset
├── teen_phone_addiction_dataset.csv # Raw dataset
├── requirements.txt
└── templates/
    ├── base.html                   # Base layout (nav, footer)
    ├── index.html                  # Homepage with stats
    ├── predict.html                # Prediction form
    ├── dashboard.html              # Analytics charts
    └── about.html                  # Project info

Run the app

python app.py

Then open: http://localhost:5000

Pages

• / → Homepage with dataset statistics
• /predict → AI prediction form (21 features → daily usage hours)
• /dashboard → Interactive charts and analytics
• /about → Project & model details

Model

• Algorithm: KNN Regressor (k=2)
• Features: 21 behavioral & psychological
• Target: Daily_Usage_Hours
• Risk Levels: Low / Moderate / High / Critical