{"title":"Assessing the Quality of Biomedical Boolean Search Strings Generated by Prompted and Unprompted Models Using ChatGPT: A Pilot Study.","authors":"Robyn B Reed, Derek J Barnett","doi":"10.1080/02763869.2024.2440848","DOIUrl":null,"url":null,"abstract":"<p><p>This pilot study investigated the use of Generative AI using ChatGPT to produce Boolean search strings to query PubMed. The goals were to determine if ChatGPT could be used in search string formation and if so, which approach was most effective. Research outputs from published systematic reviews were compared to outputs from AI generated search strings. While moderate overlap in publication retrieval between published and AI generated search strings was noted, the numbers were not sufficient to completely replicate published search strings and little difference was observed between prompted and unprompted GPT in using ChatGPT.</p>","PeriodicalId":39720,"journal":{"name":"Medical Reference Services Quarterly","volume":" ","pages":"1-10"},"PeriodicalIF":0.0000,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical Reference Services Quarterly","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/02763869.2024.2440848","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Social Sciences","Score":null,"Total":0}
引用次数: 0
Abstract
This pilot study investigated the use of Generative AI using ChatGPT to produce Boolean search strings to query PubMed. The goals were to determine if ChatGPT could be used in search string formation and if so, which approach was most effective. Research outputs from published systematic reviews were compared to outputs from AI generated search strings. While moderate overlap in publication retrieval between published and AI generated search strings was noted, the numbers were not sufficient to completely replicate published search strings and little difference was observed between prompted and unprompted GPT in using ChatGPT.
期刊介绍:
This highly acclaimed, peer-reviewed journal is an essential working tool for medical and health sciences librarians. For those professionals who provide reference and public services to health sciences personnel in clinical, educational, or research settings, Medical Reference Services Quarterly covers topics of current interest and practical value in the areas of reference in medicine and related specialties, the biomedical sciences, nursing, and allied health. This exciting and comprehensive resource regularly publishes brief practice-oriented articles relating to medical reference services, with an emphasis on user education, database searching, and electronic information. Two columns feature the Internet and informatics education.