Analyze your LINE chat conversations with beautiful visualizations and insights
This project offers two versions to suit different needs:
- ✨ Modern responsive web interface
- 🌏 Full Thai language support
- 📊 10+ interactive visualizations with cumulative toggle
- 🤝 Relationship dynamics analysis with balance scoring
- 💬 Conversation flow tracking with streak detection
- ⏰ Temporal pattern analysis (weekend/weekday, trends, seasons)
- 🏆 Engagement scoring system (0-100 with rankings)
- 🎯 Advanced analytics (response patterns, emoji usage, text prediction)
- 🚀 Better stability and performance
- 📱 Mobile-friendly design
- 💾 Large file support (up to 100MB)
- 🤖 TensorFlow-based ML predictions
- 📈 Classic Streamlit interface
- 🎯 Core analytics features
- 📊 Basic visualizations
# Clone the repository
git clone https://github.com/pannapann/line-chat-analytics.git
# Navigate to Flask version
cd line-chat-analytics/v2-flask
# Install dependencies
pip install -r requirements.txt
# Run the application
python flask_app.py
# Open http://localhost:8888 in your browser- Open LINE app on your mobile device
- Select the chat you want to analyze
- Tap settings (≡ menu → Settings)
- Choose "Export chat history"
- Select "Without media" for faster processing
- Save the .txt file and upload to the app
| Feature | v1 (Streamlit) | v2 (Flask) |
|---|---|---|
| 📊 Basic Analytics | ✅ | ✅ Enhanced |
| 🎨 Interactive Charts | ✅ Basic | ✅ 7+ Chart Types |
| 🌏 Thai Language | ✅ Full Support | |
| 📱 Mobile Friendly | ❌ | ✅ Responsive |
| 🤖 Text Prediction | ✅ TensorFlow | ✅ Lightweight |
| 💾 Large Files | ❌ Limited | ✅ Up to 100MB |
| 📈 Advanced Stats | ❌ | ✅ Response patterns, emoji analysis |
| 🎯 Easy Setup | ✅ Simple |
- 📈 Message Volume: Daily/hourly patterns, trends over time
- 👥 User Activity: Individual statistics, most active users
- 💬 Response Patterns: Average response times, conversation starters
- 📝 Text Analysis: Word frequency, message length analysis
- 😊 Emoji Usage: Most used emojis by user
- 🔮 Text Prediction: Predict message author from text content
- 🌈 Word Clouds: Beautiful visualizations (Thai + English fonts)
- 📊 Interactive Charts: Zoom, filter, and explore your data
- Backend: Flask 2.3+, pandas, plotly
- UI: Bootstrap 5, Font Awesome, custom CSS
- Thai Support: pythainlp for proper text processing
- Charts: Plotly.js for interactive visualizations
- File Handling: Up to 100MB chat exports
- Backend: Streamlit 1.12, TensorFlow 2.x
- ML: TensorFlow/Keras for text prediction
- Limitations: Dependency conflicts, smaller file limits
Found a bug or want to add features? Contributions are welcome!
- Fork the repository
- Create a feature branch
- Make your changes
- Submit a pull request
This project is open source under the MIT License.
- 🎨 Complete UI redesign with LINE-themed colors
- 🌏 Native Thai language support with proper tokenization
- 📊 7 new chart types including user comparisons and time patterns
- 🔮 Improved text prediction without heavy ML dependencies
- 📱 Mobile-responsive design that works on all devices
- ⚡ Better performance with optimized data processing
- 📈 Advanced analytics including response patterns and emoji usage
- 🛠️ Easier deployment with simplified dependencies
Start exploring your LINE chat data today! 🚀

