license: odc-by
task_categories:
- text-classification
- token-classification
- table-question-answering
- question-answering
- zero-shot-classification
- summarization
- feature-extraction
- text2text-generation
language:
- en
tags:
- medical
- climate
- code
- chemistry
- biology
- finance
- legal
pretty_name: Openalex Jun 2025 Snapshot
size_categories:
- 10M<n<100M
🎓 OpenAlex: The World's Scholarly Knowledge Graph
174M scholarly works from the world's largest open bibliographic database
OpenAlex Homepage: https://openalex.org
API Documentation: https://docs.openalex.org
Paper: https://arxiv.org/abs/2205.01833
What is OpenAlex?
🎓 OpenAlex is a free and open catalog of the global research system, containing metadata for 250M+ scholarly works, 90M+ authors, 120K+ venues, and 100K+ institutions. Named after the ancient Library of Alexandria, it serves as the successor to Microsoft Academic Graph.
This Hugging Face dataset provides a streamlined subset of 174M scholarly works with rich metadata, making it easy to:
- 📊 Analyze research trends across disciplines
- 🔗 Build citation networks and knowledge graphs
- 🤖 Train models for academic text understanding
- 🔍 Develop scholarly search and recommendation systems
Dataset Overview
Quick Stats
- 174M scholarly works (papers, books, datasets, etc.)
- Multi-disciplinary coverage: Medicine, Biology, Chemistry, Physics, Computer Science, Social Sciences, and more
- Rich metadata: Titles, abstracts, authors, institutions, citations, concepts, and open access status
- Time span: Publications from 1800s to present
- Languages: Primarily English, with multilingual content
Key Features
✅ Comprehensive Metadata
- Unique identifiers (DOI, OpenAlex ID, arXiv, PubMed)
- Publication details (title, date, venue, language)
- Author and institutional affiliations
- Citation counts and referenced works
✅ Open Access Information
- OA status and licensing
- Repository locations
- Full-text availability
✅ Concept Tagging
- Hierarchical subject classification
- Auto-generated topic tags
- Cross-disciplinary connections
✅ Quality Indicators
- Publication types
- Peer review status
- Citation metrics
Usage
Loading the Dataset
from datasets import load_dataset
# Load a sample (recommended for exploration)
dataset = load_dataset("sumuks/openalex", split="train", streaming=True)
# Iterate through examples
for paper in dataset.take(100):
print(f"Title: {paper['title']}")
print(f"Year: {paper['publication_year']}")
print(f"Citations: {paper['cited_by_count']}")
print("---")
License
This dataset is released under the Open Data Commons Attribution License (ODC-By) v1.0, following OpenAlex's commitment to open science. You are free to share and adapt this data with proper attribution.
Citation
If you use this dataset, please cite:
@article{priem2022openalex,
title={OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts},
author={Priem, Jason and Piwowar, Heather and Orr, Richard},
journal={arXiv preprint arXiv:2205.01833},
year={2022}
}
Acknowledgments
Special thanks to:
- The OurResearch team for creating and maintaining OpenAlex
- Arcadia Fund for funding OpenAlex development
- The academic community for supporting open science initiatives
Questions or Issues? Open a discussion in the Community tab or visit OpenAlex Help