OpenAIRE Research Graph Scraper

Search and export the OpenAIRE Research Graph through its official cursor API. One Actor covers research products, funded projects, research organizations, repositories/data sources and researcher profiles, producing analysis-ready records for commercial and academic workflows.

Buyer use cases

• R&D and competitor intelligence
• Grant, funder and consortium discovery
• University and research-organization enrichment
• Pharma/biotech landscape analysis
• Scholarly search, bibliometrics and citation datasets
• AI/RAG corpora with authors, subjects, identifiers and provenance

OpenAIRE spans a very large research corpus. Cursor pagination bypasses the normal 10,000-result page window and lets a run save up to 500,000 unique records.

Input

Field	Description	Default
entityType	researchProducts, projects, organizations, dataSources, or persons	researchProducts
query	Full-text OpenAIRE search	artificial intelligence
countryCodes	Optional two-letter country filters	all
types	Optional exact type labels	all
fromYear / toYear	Publication/start-year interval	none
openAccessOnly	Keep open/green records only	false
maxResults	Maximum unique records, 1–500,000	500
includeRawRecord	Include the full nested API response	false

Example — recent open-access climate research:

{
  "entityType": "researchProducts",
  "query": "climate change",
  "fromYear": 2024,
  "openAccessOnly": true,
  "maxResults": 100000
}

Example — AI-funded projects:

{
  "entityType": "projects",
  "query": "artificial intelligence",
  "fromYear": 2020,
  "maxResults": 50000
}

Output

The normalized union schema exposes ID, entity type, title/name, project/source code, type, dates/year, description, authors, countries, language, publisher, access right, citation count, persistent identifiers, subjects, website, funded amount/total cost/currency, funding programmes, projects and participating organizations. Enable includeRawRecord when you need the complete entity-specific API response; leaving it disabled substantially improves bulk throughput.

{
  "id": "openaire____::example",
  "entityType": "researchProducts",
  "title": "Example research article",
  "type": "publication",
  "year": 2025,
  "authors": ["Ada Example", "Linus Researcher"],
  "accessRight": "OPEN",
  "citationCount": 12,
  "identifiers": [{"scheme":"doi","value":"10.0000/example"}]
}

API usage

curl -X POST "https://api.apify.com/v2/acts/YOUR_USERNAME~openaire-research-graph-scraper/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"entityType":"organizations","query":"quantum","maxResults":10000}'

Source and limitations

Records come from the OpenAIRE Graph API. OpenAIRE aggregates multiple providers, so field availability, deduplication, access labels, citation metrics and update times vary. Local filters apply to the records returned by OpenAIRE's relevance-ordered search; verify critical claims with the linked identifiers/providers.

Changelog

• 2026-08-01 — v1.0.0: Initial release. Added five graph entity types, resilient cursor pagination, local country/type/year/open-access filters, deduplication, normalized plus optional raw output, 500,000-item runs and live default/multi-entity tests.