Classroom analytics with educational big data: A comparative approach for sustainable teacher reflection.

Saved in:

Bibliographic Details
Title:	Classroom analytics with educational big data: A comparative approach for sustainable teacher reflection.
Authors:	Nakamura, Kohei^1,2 nakamura-k15@cc.osaka-kyoiku.ac.jp, Horikoshi, Izumi³ horikoshi.izumi.7f@kyotou.ac.jp, Okumura, Koki² okumura.kouki.27m@st.kyoto-u.ac.jp, Ogata, Hiroaki³ hiroaki.ogata@gmail.com
Source:	Educational Technology & Society. Jul2026, Vol. 29 Issue 3, p114-130. 17p.
Subject Terms:	Reflective teaching, Student engagement, *Teaching aids, Radial basis functions, Cosine function, Data mining, Time series analysis
Abstract:	This study investigates the use of educational big data to support classroom analytics, focusing on extracting comparable classroom datasets to enhance teacher reflection. We developed methods to automatically identify and analyze comparable classroom datasets based on student interactions with digital learning materials using log data. Through two analyses, we examined the effectiveness of cosine similarity, the radial basis function (RBF) kernel, and the Jaccard coefficient for identifying comparable classes. Additionally, Dynamic Time Warping (DTW) was used to evaluate classroom engagement patterns. Our results indicate that cosine similarity and the RBF kernel are effective for detecting similarities in classroom data, while the Jaccard coefficient is less dependable. The classroom engagement analysis showed significant differences in engagement patterns across specific classes, offering opportunities for teachers to reflect on and improve their teaching practices. This approach offers a scalable way to gather data and feedback continuously, encouraging ongoing reflections on teaching strategies. Future research should apply these methods in various educational environments to validate their effectiveness. [ABSTRACT FROM AUTHOR]
	Copyright of Educational Technology & Society is the property of International Forum of Educational Technology & Society (IFETS) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Education Research Complete

Description
Abstract:	This study investigates the use of educational big data to support classroom analytics, focusing on extracting comparable classroom datasets to enhance teacher reflection. We developed methods to automatically identify and analyze comparable classroom datasets based on student interactions with digital learning materials using log data. Through two analyses, we examined the effectiveness of cosine similarity, the radial basis function (RBF) kernel, and the Jaccard coefficient for identifying comparable classes. Additionally, Dynamic Time Warping (DTW) was used to evaluate classroom engagement patterns. Our results indicate that cosine similarity and the RBF kernel are effective for detecting similarities in classroom data, while the Jaccard coefficient is less dependable. The classroom engagement analysis showed significant differences in engagement patterns across specific classes, offering opportunities for teachers to reflect on and improve their teaching practices. This approach offers a scalable way to gather data and feedback continuously, encouraging ongoing reflections on teaching strategies. Future research should apply these methods in various educational environments to validate their effectiveness. [ABSTRACT FROM AUTHOR]
ISSN:	11763647
DOI:	10.30191/ETS.202607_29(3).RP07