CrossRef Academic Scraper — DOI, Citation & Paper Metadata Extractor

Extract academic paper metadata, DOIs, authors, citations, and abstracts from CrossRef via the official public REST API. This Actor is designed for researchers, data scientists, bibliometricians, and anyone who needs structured academic publication data at scale.

CrossRef is the largest open scholarly metadata registry, indexing over 150 million works. This Actor queries that registry directly — no web scraping, no browser overhead, no proxy costs. Results are fast, clean, and reliable.

Use Cases

• Literature review automation — Build structured datasets of papers on any topic for systematic reviews or meta-analyses.
• Bibliometric research — Collect citation counts, reference networks, and publisher distributions across disciplines.
• Journal quality analysis — Filter by journal, publisher, or publication type to map the scholarly landscape.
• Research trend tracking — Monitor emerging topics by querying new publications year over year.
• Reference management — Export paper metadata with DOIs and URLs for import into Zotero, Mendeley, or EndNote.

Input

Field	Type	Required	Description
searchQuery	String	Yes	Search terms to find academic works (e.g., "machine learning", "climate change")
maxItems	Integer	No	Maximum results to extract, 1–1000 (default: 100)
workType	String	No	Filter by type: journal-article, book-chapter, proceedings-article, book, dataset, etc.
publisher	String	No	Filter by publisher name (partial match)
fromYear	Integer	No	Only include works published on or after this year
toYear	Integer	No	Only include works published on or before this year
sort	String	No	Sort by relevance, published, updated, or deposited (default: relevance)
proxyConfiguration	Object	No	Proxy settings (optional, API calls rarely need a proxy)

Output

The Actor outputs a dataset where each item represents one academic work:

{
  "title": "Machine learning vs. neutrosophic machine learning",
  "doi": "10.1201/9781003606055-5",
  "url": "https://doi.org/10.1201/9781003606055-5",
  "type": "book-chapter",
  "publisher": "CRC Press",
  "publishedDate": "2025-03-26",
  "authors": ["Florentin Smarandache"],
  "abstract": "",
  "subjects": ["Computer Science", "Artificial Intelligence"],
  "referenceCount": 0,
  "citationCount": 12,
  "containerTitle": "Neutrosophic Machine Learning",
  "page": "",
  "volume": "",
  "issue": "",
  "issn": ""
}

All fields are consistently typed. Missing values return empty strings or zeroes rather than nulls, making downstream processing easier.

Pricing

Pay per event: $0.001 per extracted work.

Because this Actor uses the public CrossRef REST API directly, there are no proxy or browser costs. Margins exceed 95%, allowing us to pass those savings on to you at a highly competitive rate.

Limitations

• CrossRef abstracts are not always available; many works return an empty abstract field.
• Author names are formatted as provided by publishers and may vary in completeness.
• CrossRef rate limits apply; the Actor handles 429 responses with automatic retry and exponential backoff.
• The API returns a maximum of 1,000 results per query due to CrossRef pagination limits.

FAQ

Q: Do I need an API key? A: No. CrossRef's public API does not require authentication for read operations.

Q: Can I filter by specific journals? A: Yes. Use the publisher filter or post-process the containerTitle field in your downstream pipeline.

Q: Are citation counts real-time? A: Citation counts are updated periodically by CrossRef and reflect their latest deposited data.

Changelog

• v1.0.0 — Initial release with search, filtering, author extraction, and citation counts.