barissozudogru/openalex-concepts

OpenAlex L1 + L2 Concepts

What this is

A snapshot of OpenAlex's Level 1 (broad fields) and Level 2 (subfields) concepts: 21,739 records across 284 broad fields and 21,455 subfields.

Level	Count	Examples
1	284	Computer science, Physics, Biology, Sociology
2	21,455	Machine learning, Quantum mechanics, Convolutional neural network

Each record:

Column	Type	Description
id	string	OpenAlex concept ID (e.g., C121955636)
name	string	Concept display name
level	int	1 (broad field) or 2 (subfield)

Source

Derived from OpenAlex, filtered to level:1|2. OpenAlex publishes its data under CC0 1.0, so this derivative is also CC0.

Load it

from datasets import load_dataset
ds = load_dataset("barissozudogru/openalex-concepts")
print(ds["train"][0])
# {'id': 'C121955636', 'name': 'Accounting', 'level': 1}

Or directly with pandas:

import pandas as pd
df = pd.read_csv("hf://datasets/barissozudogru/openalex-concepts/concepts.csv")
print(df[df["level"] == 1].head())

Use cases

• Topic classification or tagging for academic papers
• Vocabulary for concept-aware retrieval and federated search
• Building lookup tables from OpenAlex concept IDs to display names
• Training topic-classification models with a stable, offline-usable taxonomy

Caveats

• This is a snapshot, not a live mirror. OpenAlex's concept tree is updated periodically; rebuild from the source if you need the latest.
• OpenAlex deprecated their concepts API in favor of topics in late 2024; concepts remain available but are no longer the primary OpenAlex tagging system. Use accordingly.
• IDs are stable per concept but the wider OpenAlex taxonomy may reorganize over time.

Related

• research-papers-mcp — federated academic search MCP server that uses this vocabulary
• Live demo — federated search across arXiv, PubMed, Semantic Scholar, OpenAlex
• OpenAlex API documentation
• OpenAlex concepts reference (deprecated but still available)