A newer version of the Gradio SDK is available:
5.44.1
metadata
title: BIST DP-LSTM Trading Dashboard
emoji: 🚀
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 5.44.0
app_file: app.py
pinned: false
license: mit
🚀 BIST DP-LSTM Trading Dashboard
Advanced stock trading signals dashboard for BIST (Borsa Istanbul) using Differential Privacy LSTM models and Turkish financial sentiment analysis.
🎯 Features
- 📊 Interactive Price Charts with trading signals visualization
- 🤖 DP-LSTM Ensemble Models for price prediction
- 💭 Turkish Financial Sentiment Analysis from news sources
- 📈 Performance Metrics and trading statistics
- 🔒 Differential Privacy for data protection
- ⚡ Real-time Analysis (demo mode with mock data)
🛠️ Technology Stack
- Models: Differential Privacy LSTM + Temporal Fusion Transformer + Simple Transformer
- NLP: Turkish VADER sentiment + FinBERT
- Features: 131+ technical indicators across multiple timeframes
- Privacy: Adaptive differential privacy with Opacus
- Frontend: Gradio interactive dashboard
- Deployment: Hugging Face Spaces
📊 Performance Targets
- Direction Accuracy: ≥68% (MVP) → ≥75% (Production)
- Sharpe Ratio: >2.0
- Max Drawdown: <15%
- Signal Confidence: >65% average
⚠️ Disclaimer
This system is for research and educational purposes only.
- Past performance does not guarantee future results
- Not financial advice - consult with financial advisors
- Demo mode shows simulated data for demonstration
- Always do your own research before investing
🔗 Links
- GitHub Repository: RSMCTN/BIST_AI001
- Production API: Railway Deployment
- Documentation: Complete System Docs
🎮 How to Use
- Select a BIST 30 stock from the dropdown
- Choose analysis period (7-90 days)
- Click "Analyze Stock" button
- Explore interactive charts and metrics
- Review latest trading signals and performance
🚀 About the System
This dashboard showcases an advanced algorithmic trading system designed for the Turkish stock market (BIST). The system integrates:
- Multi-source data collection (MatriksIQ API, financial news)
- Advanced ML models with privacy protection
- Risk management and portfolio optimization
- Real-time execution capabilities
- Comprehensive monitoring and analytics
Built with modern MLOps practices and deployed across multiple platforms for maximum accessibility and reliability.
Made with ❤️ for the Turkish financial research community