Dataset
clinical-trials-xml-2018-2024
ClinicalTrials.gov XML for studies registered between 2018 and 2024, parsed into CSV, hundreds of thousands of records under CC0. A clean, licensed point-in-time history of US clinical-trial registrations.
Data preview
Dataset card
clinical-trials-xml-2018-2024 is a Hugging Face dataset published by dmariko that consolidates PubMed-style article metadata with extracted paragraph text from referenced publications, distributed under a CC0 1.0 tag. It contains a single training split of roughly 628,000 rows in tabular form.
What this dataset contains
The dataset joins bibliographic records for biomedical articles with paragraph-level text extracted from referenced figures, tables and supplementary materials, identified via columns such as AccessionID, RefID and PMID_x. Each row pairs a citation (Title, Authors, Journal/Book, Publication Year, DOI, NIHMS ID) with the underlying paragraph snippet, supporting fine-grained linking between claims and source literature. A Retracted flag is included so that downstream users can filter out withdrawn studies.
Structure and fields
The single CSV-distributed training split holds 627,629 rows and mixes string identifiers with one numeric year column (Publication Year) and a numeric PMID_y field. Identifier columns include PMID_x, PMCID_x, RefID, AccessionID, PMID_y, PMCID_y, DOI and NIHMS ID, while content fields carry paragraph text, subjects, keywords, Citation, First Author and Article Citation. Operational columns such as Key, ETag, Last Updated UTC (YYYY-MM-DD HH:MM:SS), License and Retracted support synchronization and provenance tracking with the upstream repository.
How analysts use it
Analysts can join paragraph text to bibliographic metadata to build retrieval, summarization or citation-graph datasets for biomedical NLP research, filtering with the Retracted field to keep only valid literature. The CC0 1.0 tag and CSV/tabular format make it compatible with pandas, polars and the mlcroissant library for reproducible loading. Use cases supported by the schema include evidence-linking for systematic reviews and constructing training corpora for scientific question answering, rather than investment analysis.
Documentation & references 1
Related datasets
Browse moreParsed eligibility criteria from 13,229 ClinicalTrials.gov studies representing the candidate pool gathered by two first-stage retrievers during TREC Clinical Trials 2021–2023, formatted as typed entity-relation graphs for reranking models.
The clinical-trials-patient-graphs dataset is a small tabular collection of patient-level clinical records distributed in Parquet format by the 2001jdev publisher. It organizes entities and relations extracted from trial data into structured rows spanning 2021 through 2023.
This dataset is a parsed snapshot of ClinicalTrials.gov records, distributed in Parquet format on the Hugging Face Hub under an unspecified license. It contains roughly 52,000 trial entries drawn from public registry filings.
clinical-trials-trec-qrels is a tabular relevance-judgment file distributed by the 2001jdev user on the Hugging Face Hub. It maps clinical-trial topic identifiers to NCT registry IDs with graded relevance scores.
The clinical-trials-trec-topics dataset is a small tabular collection of clinical-trial topic records distributed in Parquet format on the Hugging Face Hub. It appears to be derived from TREC clinical-trial retrieval topics, with rows partitioned by year.
SEC EDGAR 10-K & 10-Q financial statements for 2,600+ NASDAQ companies.
Discussion
Questions or observations about this dataset? Join the conversation on X.
No comments yet. Start the conversation on X.