VitaLITy 2.0 - REST API

Backend API server for VitaLITy, built with Flask, ChromaDB, and LangChain.

---

Requirements

• Python 3.9+
• Azure OpenAI API access (for LLM features)

---

Setup

1. Create Virtual Environment

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

2. Install Dependencies

pip install --upgrade pip
pip install -r requirements.txt

3. Configure Environment Variables

Create a .env file in the project root:

# Azure OpenAI (required for LLM features)
AZURE_OPENAI_ENDPOINT=https://your-endpoint.openai.azure.com/
AZURE_OPENAI_API_VERSION=2025-04-01-preview
AZURE_OPENAI_DEPLOYMENT=your-deployment-name
AZURE_OPENAI_API_KEY=your-api-key

# Azure OpenAI Embeddings
AZURE_OPENAI_EMBED_DEPLOYMENT=text-embedding-3-small
AZURE_OPENAI_EMBED_API_VERSION=2024-02-01

# Optional: Semantic Scholar API
SEMANTIC_SCHOLAR_API_KEY=your-key

4. Load Data into ChromaDB

python load_to_chroma.py

This creates the vector database in chroma_db/ from data/VitaLITy-2.0.0_final.json.

---

Run

Development

# Run in production mode (default)
python main_chroma.py

# Run in debug mode with auto-reload
python main_chroma.py --debug

The API will be available at http://localhost:3000 (or the port specified in the PORT environment variable).

Note: Debug mode enables Flask's debug mode and auto-reloader, which automatically restarts the server when code changes are detected. Use --debug only during development.

Production (with Gunicorn)

pip install gunicorn eventlet
gunicorn --worker-class eventlet -w 1 --bind 127.0.0.1:8000 --timeout 600 main_chroma:app

---

API Endpoints

Paper Retrieval

Endpoint	Method	Description
/getPapers	POST	Get papers by IDs or titles
/getSimilarPapers	POST	Find similar papers by embedding
/getSimilarPapersByAbstract	POST	Find similar papers by abstract text
/getUmapPoints	GET	Get 2D UMAP coordinates for visualization
/getMetaData	GET	Get metadata for UI filters

LLM Features

Endpoint	Method	Description
/chat	POST	Chat with LLM about selected papers
/summarize	POST	Summarize selected papers
/literatureReview	POST	Generate a literature review

Example Requests

Get Similar Papers:

POST /getSimilarPapers
{
    "input_data": ["Paper Title 1", "Paper Title 2"],
    "input_type": "Title",
    "limit": 10,
    "embedding": "ada",
    "dimensions": "2D"
}

Find Papers by Abstract:

POST /getSimilarPapersByAbstract
{
    "input_data": {
        "title": "My Paper Title",
        "abstract": "This paper explores..."
    },
    "limit": 10
}

Chat with Papers:

POST /chat
{
    "papers": [...],
    "message": "What are the main themes in these papers?"
}

---

Project Structure

rest-api-main/
├── main_chroma.py      # Main Flask application
├── config.py           # Configuration settings
├── load_to_chroma.py   # Script to load data into ChromaDB
├── prompt.py           # LLM prompts
├── requirements.txt    # Python dependencies
├── .env                # Environment variables (not committed)
├── data/
│   └── VitaLITy-2.0.0_final.json  # Paper data with embeddings
├── chroma_db/          # ChromaDB vector database
├── service/            # Query and RAG functions
├── model/              # Constants and schemas
└── extension/          # Data caching

---

Deployment

See DEPLOYMENT.md for VM deployment instructions.

---

Credits

VitaLITy was created by Arpit Narechania, Alireza Karduni, Ryan Wesslen, and Emily Wall.

---

Citation

@article{narechania2021vitality,
  title={vitaLITy: Promoting Serendipitous Discovery of Academic Literature with Transformers \& Visual Analytics},
  author={Narechania, Arpit and Karduni, Alireza and Wesslen, Ryan and Wall, Emily},
  journal={IEEE Transactions on Visualization and Computer Graphics},
  year={2022},
  doi={10.1109/TVCG.2021.3114820},
  publisher={IEEE}
}

---

License

The software is available under the MIT License.

---

Contact

If you have any questions, feel free to open an issue or contact Arpit Narechania.