OpenAlex Scholarly Data Extractor

Extract scholarly data at scale from OpenAlex — works (papers), authors, institutions, journals/sources, publishers, funders, topics, and concepts — one record per row.

Built for research-intelligence and R&D competitive analysis, bibliometrics, science-funding analysis, and academic data teams.

---

Why use this actor

• Massive corpus — 316M+ works (verified), plus ~95M authors and the full institution/journal/funder graph.
• One record per row, with a flat header (openalex_id, display_name, cited_by_count, publication_year) plus the full raw OpenAlex object (authorships, institutions, open-access, IDs, …).
• No login, no key. Add your email to join the faster "polite pool".

Input

Field	Type	Description
entity	dropdown	Works / Authors / Sources / Institutions / Publishers / Funders / Topics / Concepts.
filter	text	OpenAlex filter, e.g. publication_year:2026,is_oa:true.
search	text	Full-text keyword search.
mailto	text	Your email — joins OpenAlex's faster pool (recommended).
perPage	int	1–200 (default 200).
maxItems	int	0 = all matching (up to the 100k cursor window).

entity is a pick-list. filter/search are free-text because OpenAlex's filter grammar is open-ended (documented at docs.openalex.org).

Output

Envelope + recordType (e.g. WORK, AUTHOR) + entity_type + flat header, then the raw OpenAlex object:

{
  "_input": "filter=publication_year:2026,is_oa:true",
  "_source": "S1-openalex",
  "_scrapedAt": "2026-06-03T10:00:00Z",
  "recordType": "WORK",
  "entity_type": "works",
  "openalex_id": "https://openalex.org/W...",
  "display_name": "...",
  "cited_by_count": 12,
  "publication_year": 2026,
  "doi": "https://doi.org/...",
  "authorships": [ "..." ],
  "open_access": { "is_oa": true, "...": "..." }
}

How it works

1. Your entity choice and filters define what to pull.
2. The actor automatically pages through all matching results (up to 200 per request).
3. Each entity streams into the dataset.

Known limits

• Public data — no account needed, runs from any connection. Backs off on HTTP 429. Add mailto for the faster pool.
• A single query streams up to ~100,000 results; to pull a larger slice, narrow with filter (e.g. by year) and run per slice.
• Verified live 2026-06-03: works total 315,965,530; cursor=* → meta.next_cursor confirmed.