wshuai190/pubmed-pmc-sr-filtered

Dataset Description

This dataset contains medical literature data for training Boolean query generation models. The data includes PubMed articles with their associated metadata, references, and result section PMIDs.

Dataset Structure

Data Fields

• pmid: PubMed ID of the article
• pmc-id: PMC ID (if available)
• title: Article title
• max-date: Maximum publication date
• references-pmids: List of PMIDs referenced in the article
• first_ref_date: Date of first reference
• last_ref_date: Date of last reference
• results-section-pmids: List of PMIDs found in the results section
• topicid: Topic identifier (defaults to pmid if not present)

Data Splits

• Train: 32,794 samples - Earlier publications for training
• Test: 32,794 samples - Later publications for evaluation
• Temporal1000: 1,000 samples - Recent publications (after Nov 1, 2024) for temporal evaluation

Usage

from datasets import load_dataset

# Load the full dataset
dataset = load_dataset("wshuai190/pubmed-pmc-sr-filtered")

# Load specific split
train_data = load_dataset("wshuai190/pubmed-pmc-sr-filtered", split="train")
test_data = load_dataset("wshuai190/pubmed-pmc-sr-filtered", split="test")
temporal_data = load_dataset("wshuai190/pubmed-pmc-sr-filtered", split="temporal1000")

Citation

@inproceedings{autobool2026,
  title={AutoBool: Reinforcement Learning for Boolean Query Generation in Systematic Reviews},
  author={[Shuai Wang, Harrisen Scells, Bevan Koopman, Guido Zuccon]},
  booktitle={Proceedings of the 2026 Conference of the European Chapter of the Association for Computational Linguistics (EACL)},
  year={2025}
}