Show simple item record

dc.contributor.authorLämsä, Joni
dc.contributor.authorUribe, Pablo
dc.contributor.authorJiménez, Abelino
dc.contributor.authorCaballero, Daniela
dc.contributor.authorHämäläinen, Raija
dc.contributor.authorAraya, Roberto
dc.date.accessioned2021-04-12T07:38:31Z
dc.date.available2021-04-12T07:38:31Z
dc.date.issued2021
dc.identifier.citationLämsä, J., Uribe, P., Jiménez, A., Caballero, D., Hämäläinen, R., & Araya, R. (2021). Deep Networks for Collaboration Analytics : Promoting Automatic Analysis of Face-to-Face Interaction in the Context of Inquiry-Based Learning. <i>Journal of Learning Analytics</i>, <i>8</i>(1), 113-125. <a href="https://doi.org/10.18608/jla.2021.7118" target="_blank">https://doi.org/10.18608/jla.2021.7118</a>
dc.identifier.otherCONVID_66359641
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/75023
dc.description.abstractScholars have applied automatic content analysis to study computer-mediated communication in computer-supported collaborative learning (CSCL). Since CSCL also takes place in face-to-face interactions, we studied the automatic coding accuracy of manually transcribed face-to-face communication. We conducted our study in an authentic higher-education physics context where computer-supported collaborative inquiry-based learning (CSCIL) is a popular pedagogical approach. Since learners’ needs for support in CSCIL vary in the different inquiry phases (orientation, conceptualization, investigation, conclusion, and discussion), we studied, first, how the coding accuracy of five computational models (based on word embeddings and deep neural networks with attention layers) differed in the various inquiry-based learning (IBL) phases when compared to human coding. Second, we investigated how the different features of the best performing computational model improved the coding accuracy. The study indicated that the accuracy of the best performing computational model (differentiated attention with pre-trained static embeddings) was slightly better than that of the human coder (58.9% vs. 54.3%). We also found that considering the previous and following utterances, as well as the relative position of the utterance, improved the model’s accuracy. Our method illustrates how computational models can be trained for specific purposes (e.g., to code IBL phases) with small data sets by using pre-trained models.en
dc.format.mimetypeapplication/pdf
dc.languageeng
dc.language.isoeng
dc.publisherSociety for Learning Analytics Research
dc.relation.ispartofseriesJournal of Learning Analytics
dc.rightsCC BY-NC-ND 4.0
dc.subject.othercollaboration analytics
dc.subject.othercomputational models
dc.subject.othercomputer-supported collaborative learning
dc.subject.otherCSCL
dc.subject.otherCSCIL
dc.subject.otherdeep networks
dc.subject.otherinquiry-based learning
dc.subject.otherword embedding
dc.titleDeep Networks for Collaboration Analytics : Promoting Automatic Analysis of Face-to-Face Interaction in the Context of Inquiry-Based Learning
dc.typearticle
dc.identifier.urnURN:NBN:fi:jyu-202104122335
dc.contributor.laitosKasvatustieteiden laitosfi
dc.contributor.laitosDepartment of Educationen
dc.contributor.oppiaineKasvatustiedefi
dc.contributor.oppiaineEducationen
dc.type.urihttp://purl.org/eprint/type/JournalArticle
dc.description.reviewstatuspeerReviewed
dc.format.pagerange113-125
dc.relation.issn1929-7750
dc.relation.numberinseries1
dc.relation.volume8
dc.type.versionpublishedVersion
dc.rights.copyright© 2021 the Authors
dc.rights.accesslevelopenAccessfi
dc.relation.grantnumber292466
dc.subject.ysosisällönanalyysi
dc.subject.ysovuorovaikutus
dc.subject.ysoyhteisöllinen oppiminen
dc.subject.ysotietokoneavusteinen oppiminen
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p14612
jyx.subject.urihttp://www.yso.fi/onto/yso/p10591
jyx.subject.urihttp://www.yso.fi/onto/yso/p18727
jyx.subject.urihttp://www.yso.fi/onto/yso/p7221
dc.rights.urlhttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.relation.doi10.18608/jla.2021.7118
dc.relation.funderSuomen Akatemiafi
dc.relation.funderAcademy of Finlanden
jyx.fundingprogramProfilointi, SAfi
jyx.fundingprogramResearch profiles, AoFen
jyx.fundinginformationThe publication of this article received financial support from the Academy of Finland [grant numbers 292466 and 318095, the Multidisciplinary Research on Learning and Teaching profiles I and II of University of Jyväskylä]and ANID/PIA/Basal Funds for Excellence (FB0003).


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

CC BY-NC-ND 4.0
Except where otherwise noted, this item's license is described as CC BY-NC-ND 4.0