This ethnographic study investigates how ChatGPT meets information needs, andthe difference in functional outcomes when using ChatGPT in a reference transaction compared to a librarian. The theoretical starting point is Taylor's theories on information needs and filters, commands and questions. The study is based on empirical material consisting of observations of libraries and of ChatGPT. The empirical material is analyzed using analytical autoethnography, thick description, and Childer's system for evaluating reference transactions. The purpose of the thesis is to investigate how ChatGPT performs in reference transactions compared to librarians. The study shows that ChatGPT meets information needs in a different way than librarians. The biggest difference in functional results is that ChatGPT's responses are incorrect more frequently than the librarians. ChatGPT not only makes dubious claims, it also hallucinates and recommends books that do not exist. This thesis can be used as a basis for further research, as well as support for public libraries that want to use AI technology in reference transactions.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:umu-220742 |
Date | January 2024 |
Creators | Ramström, Cecilia |
Publisher | Umeå universitet, Sociologiska institutionen |
Source Sets | DiVA Archive at Upsalla University |
Language | Swedish |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0015 seconds