Unified Scholarly Data Lake with DOI Normalization

A 525 GB unified data lake integrates seven scholarly datasets with cross-dataset DOI normalization and 13 scientific ontologies containing 1.3 million terms. It was created by J0nasW and includes a reproducible ETL pipeline. The dataset was last updated in March 2026.

Use Cases

• Analyze citation patterns and research impact using normalized DOI fields across seven integrated datasets.
• Map scientific concepts and terminology by querying the integrated network of 13 ontologies containing 1.3 million terms.
• Train models for scholarly text classification using features from the unified, multi-source academic corpus.
• Build knowledge graphs linking publications, authors, and concepts using the structured ontology and bibliographic data.

Strengths

• Integrates seven distinct scholarly datasets into a single 525 GB resource.
• Includes a network of 13 scientific ontologies with 1.3 million standardized terms for concept mapping.

Limitations

• One major source (Semantic Scholar S2AG) is not included in the redistribution due to API restrictions, limiting completeness.
• Specific details on row counts, column names, and file formats are unavailable, hindering initial assessment.

Provenance

Source

Aggregated from seven scholarly datasets; one source (Semantic Scholar S2AG) supported by pipeline but not redistributed.

Collection Method

Integrated via a reproducible ETL pipeline with cross-dataset DOI normalization.

Freshness

Last updated March 2026.