Data and analysis code for: Criterion guided large language models as additional reviewers for melanoma guideline title and abstract screening
Year: 2026
Type: dataset
Abstract: Deidentified data and the complete R analysis pipeline for the study "Criterion-guided large language models as additional reviewers for melanoma guideline title–abstract screening". The archive contains the anonymized expert and LLM screening decisions for 60 abstracts from three German S3 melanoma guideline key questions, the two-rater consensus reference standard, the six LLM workflow outputs, the per-bundle supplementary-data CSVs, the frozen analysis outputs (results and clustered-bootstrap distribution), and the R scripts that reproduce every reported statistic and supplementary figure. See README.md for the folder layout and reproduction steps, and MANIFEST.csv for a complete file listing. License terms are described in README.md.
Author: Damm, Hendrik
Institution: Dortmund University of Applied Sciences and Arts
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7204229166