scholarAPI

A FastAPI API to pull research papers through keywords, summarize papers, and get citations

📚 ScholarSearch API

An open-source research paper discovery and summarization API using the arXiv API, built with FastAPI and powered by OpenAI or Hugging Face for natural language summarization.

---

🚀 Features

• 🔍 Search academic papers by keyword using arXiv
• 📄 Retrieve metadata: title, abstract, authors, publication date, PDF link
• 🧠 Summarize abstracts or full papers using GPT-4o or BART (pluggable)
• 📎 Format citations in BibTeX (APA/MLA support coming soon!)
• 📉 Built-in rate limiting to prevent abuse
• 🔒 (Optional) API key authentication and per-user quotas (easy to add)

---

🛠️ Installation

git clone https://github.com/your-username/scholarsearch-api.git
cd scholarsearch-api
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Create a .env file based on the example:

cp .env.example .env

Edit .env and add your OpenAI API key:

OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

---

🧪 Running the API

uvicorn main:app --reload

Visit http://localhost:8000/docs for interactive API docs (via Swagger UI).

---

🌐 API Endpoints

🔍 /api/search

Search arXiv by keyword.

Example:

GET /api/search?query=transformers&limit=5

---

🧠 /api/summarize

Summarize text using GPT.

Example:

GET /api/summarize?text=Deep%20learning%20is...&length=short

Query Params:

• text – The content to summarize
• length – short, medium, or long

---

📎 /api/citation

Generate a simple citation string.

Example:

GET /api/citation?title=AI%20and%20Robotics&authors=Alice,Bob&published=2023-05-01

---

📦 File Structure

scholarsearch-api/
├── main.py            # FastAPI app and routes
├── rarxiv.py          # arXiv search wrapper
├── summarizer.py      # Summarization logic (GPT or BART)
├── citation.py        # Citation formatting
├── requirements.txt   # Dependencies
├── .env.example       # API key config template
├── README.md          # Project documentation

---

✅ To-Do (Future Improvements)

• Add support for HuggingFace Transformers (BART/PEGASUS)
• Format citations in APA and MLA
• Add /register endpoint to generate API keys
• Add usage quotas and tracking
• Deploy to Render, Vercel, or Hugging Face Spaces

---

📄 License

This project is licensed under the MIT License.

---

🤝 Contributions

Pull requests are welcome! If you’d like to contribute:

1. Fork the repo
2. Make your changes
3. Submit a PR

Let’s build a better way to search and understand research 📚