A1 Journal article (refereed)
Conceptual network of teachers' talk : Automatic analysis and quantitative measures (2020)


Caballero, D., Pikkarainen, T., Araya, R., Viiri, J., & Espinoza, C. (2020). Conceptual network of teachers' talk : Automatic analysis and quantitative measures. FMSERA Journal, 3(1), 18-31. https://journal.fi/fmsera/article/view/79630


JYU authors or editors


Publication details

All authors or editors: Caballero, Daniela; Pikkarainen, Toni; Araya, Roberto; Viiri, Jouni; Espinoza, Catalina

Journal or series: FMSERA Journal

eISSN: 2490-158X

Publication year: 2020

Volume: 3

Issue number: 1

Pages range: 18-31

Publisher: The Finnish Matehematics and Science Education Research Association

Publication country: Finland

Publication language: English

Persistent website address: https://journal.fi/fmsera/article/view/79630

Publication open access: Openly available

Publication channel open access: Open Access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/68828


Abstract

Educational field can take advantage of the improvements of Automatic Speech Recognition (ASR), since we can apply ASR algorithms in non-ideal conditions such as real classrooms. In the context of QuIP project, we used ASR systems to translate audio from teachers’ talk into text to study conceptual networks based on what the teacher says during his/her lecture, particularly the key concepts mentioned and their temporal co-occurrence. In the present study, quantitative metrics are provided, such as centrality measures and PageRank, which can be used to analyse the conceptual networks in a broaden way. With a case-study design, two teachers’ talk are described quantitatively and qualitatively using the metrics, suggesting that PageRank could be a good metric to find differences in teachers’ talk. Finally, we discuss about the potential of this kind of analysis.


Keywords: speech recognition; classroom work; teachers; speech (phenomena)


Contributing organizations


Ministry reporting: Yes

Reporting Year: 2020

JUFO rating: 1


Last updated on 2021-09-08 at 13:45