Enrich Bibliography Data with Crossref

Bibliographies often arrive with a DOI and inconsistent text fields. Enrichment can improve linking and analysis, but it should not erase curated values or hide source conflicts. The Crossref DOI Scraper returns structured metadata for a reconciliation workflow.

Resolve identifiers in bounded batches

Submit exact DOIs and keep the input record ID outside the Actor so results can be joined back deterministically.

{
  "queries": [],
  "dois": ["10.1038/s41586-021-03819-2"],
  "maxResultsPerQuery": 20,
  "mailto": "metadata@example.org"
}

Store Crossref values in source-specific columns or a versioned source table. Compare title, authors, publisher, venue, dates, ISSNs, ISBNs, language, licenses, and URL with local values.

Route differences by importance

Normalize whitespace and DOI casing before classifying conflicts. Minor punctuation differences in titles may be safe to present consistently, while author order, publication year, or identifier conflicts deserve manual review.

Keep authors structured. ORCID is a valuable identifier where available, but its absence does not mean two equal names identify the same person. Affiliations are deposited strings and can be incomplete.

Record enrichment time, Actor run ID, Crossref indexed time, and the chosen disposition for each field. Never convert missing source metadata into an empty authoritative value that deletes local data.

Schedule periodic refreshes only where corrections matter operationally. Compare Crossref indexed timestamps first to avoid rewriting unchanged records. If a run fails or returns fewer rows than the submitted DOI count, retain the prior source version and retry unresolved identifiers. A traceable enrichment pipeline makes every displayed field attributable to either local curation or a specific Crossref snapshot.