Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/1200
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSun, Weiyi
dc.contributor.authorZhang, Jianwei
dc.contributor.authorJin, Hui
dc.contributor.authorLyu, Siwei
dc.date.accessioned2018-05-21T22:08:51Z
dc.date.accessioned2019-01-22T19:43:12Z-
dc.date.available2018-05-21T22:08:51Z
dc.date.available2019-01-22T19:43:12Z-
dc.date.issued2014-06
dc.identifier.citationSun, W., Zhang, J., Jin, H., & Lyu, S. (2014). Analyzing Online Knowledge-Building Discourse Using Probabilistic Topic Models. In Joseph L. Polman, Eleni A. Kyza, D. Kevin O'Neill, Iris Tabak, William R. Penuel, A. Susan Jurow, Kevin O'Connor, Tiffany Lee, and Laura D'Amico (Eds.). Learning and Becoming in Practice: The International Conference of the Learning Sciences (ICLS) 2014. Volume 2. Colorado, CO: International Society of the Learning Sciences, pp. 823-830.en_US
dc.identifier.urihttps://doi.dx.org/10.22318/icls2014.823
dc.identifier.urihttps://repository.isls.org//handle/1/1200-
dc.description.abstractThis exploratory study tested the use of machine learning techniques, in particular, probabilistic topic models, to conduct automated analysis of the online discourse a Grade 4 knowledge-building community that investigated optics over three months using Knowledge Forum. Using the Latent Dirchilet Allocation (LDA) model, we extracted ten distinct and semantically meaningful clusters (i.e., topics) from the online discourse, which overlapped substantially with--although did not directly map onto--the inquiry themes identified by students and inquiry thread topics identified by researchers. The LDA analysis further identified discourse entries relevant to each of the topics, with acceptable agreement achieved between the automated analysis results and the manual coding of two researchers.en_US
dc.language.isoenen_US
dc.publisherBoulder, CO: International Society of the Learning Sciencesen_US
dc.titleAnalyzing Online Knowledge-Building Discourse Using Probabilistic Topic Modelsen_US
dc.typeBook chapteren_US
Appears in Collections:ICLS2014

Files in This Item:
File SizeFormat 
823-830.pdf837.35 kBAdobe PDFView/Open


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