An integrated research assistant built with retrieval, structured data pipelines, and OpenAlex ingestion.
research-llm/
├── rag/ # Retrieval-Augmented Generation system
├── database/ # Data ingestion, processing, and storage
├── openalex/ # OpenAlex data pipeline
Core retrieval and question answering engine.
- Semantic search
- Graph-based retrieval
- Streamlit interface
- Session memory
📂 Location: /rag
📖 Details: RAG README
Handles ingestion, preprocessing, and enrichment of research data.
- PDF ingestion
- Data cleaning and normalization
- Model training utilities
- Pipeline orchestration
📂 Location: /database
📖 Details: Database README
Fetches and processes academic metadata from OpenAlex.
- Paper metadata retrieval
- Download pipelines
- Chunking and normalization
- Graph + vector ingestion
📂 Location: /openalex
📖 Details: OpenAlex README
- OpenAlex → fetch research papers
- Database → clean, structure, enrich
- RAG → retrieve + answer queries
# Clone repo
git clone https://github.com/AryanApte1408/research-llm.git
cd research-llmNavigate to a module:
cd rag
# or
cd database
# or
cd openalexFollow instructions in each module’s README.
- Python
- ChromaDB
- Neo4j
- Streamlit
- OpenAlex API
Each module is independently runnable but designed to work together as a pipeline.
This project is licensed under the MIT License. See LICENSE for the full text.
Developed and maintained by the Open Source Program Office at Syracuse University. Reach out for feedback and suggested improvements:
This project was supported as part of grants (#G2023-20946, #G-2025-79206) from the Alfred P. Sloan Foundation.