https://github.com/walterthesmart/fin-search
An AI-powered financial research tool that combines Hugging Face models with real-world financial data sources
https://github.com/walterthesmart/fin-search
cyptocurrency finance financial-data research stock-market
Last synced: about 2 months ago
JSON representation
An AI-powered financial research tool that combines Hugging Face models with real-world financial data sources
- Host: GitHub
- URL: https://github.com/walterthesmart/fin-search
- Owner: walterthesmart
- Created: 2025-02-08T16:25:45.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-02-08T17:31:24.000Z (4 months ago)
- Last Synced: 2025-02-08T18:22:28.203Z (4 months ago)
- Topics: cyptocurrency, finance, financial-data, research, stock-market
- Language: TypeScript
- Homepage:
- Size: 77.1 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AI-Powered Financial Research Tool 🚀📊
A comprehensive financial analysis platform that combines real-time market data with state-of-the-art AI models from Hugging Face to provide actionable market insights and research capabilities.
## Features
- **Real-time Market Analysis** - Integration with multiple data sources for comprehensive market coverage
- **AI-Powered Sentiment Analysis** - Advanced NLP models to analyze news and social media sentiment
- **Automated Report Generation** - Intelligent summarization of earnings calls and financial reports
- **Event Detection** - Real-time monitoring of significant market events using Named Entity Recognition
- **Interactive Dashboard** - Streamlit-based visualization of market insights and sentiment analysis
## Architecture
### Data Sources
- Market Data: Alpha Vantage, Yahoo Finance, Polygon
- News & Sentiment: NewsAPI, GDELT, Reddit/Twitter API
- Fundamentals: SEC EDGAR, Quandl, EOD Historical
- Alternative Data: CryptoCompare, Glassnode
### AI Models (Hugging Face)
- Sentiment Analysis: ProsusAI/finbert, distilroberta-finetuned-financial-news-sentiment
- Text Summarization: facebook/bart-large-cnn, google/pegasus-xsum
- Named Entity Recognition: dslim/bert-base-NER
- Time-Series Forecasting: huggingface/autonlp-timeseries-1
- Question Answering: deepset/roberta-base-squad2
## Installation
```bash
# Clone the repository
git clone https://github.com/walterthesmart/Fin-Search.git
cd Fin-Search
# Create and activate virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
```
## Quick Start
1. Set up your environment variables:
```bash
export HUGGINGFACE_API_KEY="your_key_here"
export ALPHA_VANTAGE_API_KEY="your_key_here"
export NEWS_API_KEY="your_key_here"
```
2. Run the sentiment analysis dashboard:
```bash
streamlit run app/dashboard.py
```
3. Access the dashboard at `http://localhost:8501`
## Usage Examples
### Sentiment Analysis
```python
from pipelines.sentiment import analyze_sentiment
# Analyze sentiment for a specific ticker
sentiments = analyze_sentiment("AAPL")
print(f"Sentiment Analysis Results: {sentiments}")
```
### Earnings Call Summary
```python
from pipelines.summarization import summarize_earnings_call
# Summarize an earnings call transcript
summary = summarize_earnings_call("path_to_transcript.txt")
print(f"Summary: {summary}")
```
## Project Structure
```
.
├── data/ # Historical datasets
├── models/ # Fine-tuned Hugging Face models
├── pipelines/
│ ├── sentiment.py # News sentiment analysis
│ ├── summarization.py # Earnings call summaries
│ └── forecasting.py # Time-series predictions
├── app/ # Streamlit/FastAPI frontend
├── Dockerfile # Containerization
└── requirements.txt
```
## Deployment
The application can be deployed using Docker:
```bash
# Build Docker image
docker build -t Fin-Search .
# Run container
docker run -p 8501:8501 Fin-Search
```
## Important Notes
### Challenges & Mitigations
- **Latency**: Optimized with distilled models for real-time analysis
- **Data Quality**: Combined with technical indicators for validation
- **Model Accuracy**: Implemented fact-checking layer using SEC data
- **Compliance**: Focus on research insights rather than direct trading signals
### Regulatory Compliance
This tool is designed for research purposes only. Users should:
- Consult with legal advisors before deployment
- Ensure compliance with financial regulations
- Not rely solely on this tool for trading decisions
## Contributing
1. Fork the repository
2. Create your feature branch (`git checkout -b feature/AmazingFeature`)
3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
4. Push to the branch (`git push origin feature/AmazingFeature`)
5. Open a Pull Request
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## Acknowledgments
- Hugging Face for providing state-of-the-art NLP models
- Various data providers for financial market data
- Open-source community for tools and libraries
## Contact
Your Name - [@kudiegomoney](https://twitter.com/kudiegomoney)
Project Link: [https://github.com/walterthesmart/Fin-Search](https://github.com/walterthesmartFin-Search)
![alt text]()