Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/10507
Full metadata record
DC FieldValueLanguage
dc.contributor.authorBosch, Nigel
dc.contributor.authorWilliams-Dobosz, Destiny
dc.contributor.authorPerry, Michelle
dc.date.accessioned2024-06-10T04:55:38Z-
dc.date.available2024-06-10T00:51:23Z
dc.date.available2024-06-10T04:55:38Z-
dc.date.issued2024
dc.identifier.citationBosch, N., Williams-Dobosz, D., & Perry, M. (2024). Measuring Help-seeking in Online Course Discussion Forums with Privacy-preserving Large Language Models. In Clarke-Midura, J., Kollar, I., Gu, X., & D'Angelo, C. (Eds.), Proceedings of the 17th International Conference on Computer-Supported Collaborative Learning - CSCL 2024 (pp. 189-192). International Society of the Learning Sciences.en
dc.identifier.urihttps://doi.org/10.22318/cscl2024.507823
dc.identifier.urihttps://repository.isls.org//handle/1/10507-
dc.descriptionShort Paperen
dc.description.abstractDiscussion forums in college courses play a vital role in enabling students to seek academic help from each other and from instructors, especially in online courses. However, it can also be difficult for instructors and researchers interested in help-seeking to identify when it occurs in large, active forums. This study examined large language models as measurement tools for automatically coding students’ help-seeking forum posts, resulting in Spearman’s rho up to .711 for the correlation between models’ help-seeking ratings versus manually coding via an established schema. The large language model approach requires no problem-specific training data, enabling the help-seeking model to be constructed with minimal manual coding of forum posts compared to traditional machine learning approaches. Moreover, the models in this study are offline models, run on a typical laptop, which preserve students’ privacy by eliminating the need to transmit forum data to a third-party service.en
dc.publisherInternational Society of the Learning Sciences
dc.relation.ispartofurn:ISBN:979-8-9906980-1-7
dc.relation.ispartofProceedings of the 17th International Conference on Computer-Supported Collaborative Learning - CSCL 2024, pp. 189-192
dc.subjectCSCLen
dc.titleMeasuring Help-seeking in Online Course Discussion Forums with Privacy-preserving Large Language Modelsen
dc.typeconferencePaper
dc.identifier.doi10.22318/cscl2024.507823
Appears in Collections:ISLS Annual Meeting 2024

Files in This Item:
File SizeFormat 
CSCL2024_189-192.pdf281.12 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.