Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/10375
Title: Detection of Goal Setting and Planning in Self-regulated Learning Using Machine Learning and Think-aloud Protocols
Authors: Li, Shan
Zheng, Juan
Huang, Xiaoshan
Wang, Tingting
Lajoie, Susanne P.
Keywords: Learning Sciences
Issue Date: 2023
Publisher: International Society of the Learning Sciences
Citation: Li, S., Zheng, J., Huang, X., Wang, T., & Lajoie, S. P. (2023). Detection of goal setting and planning in self-regulated learning using machine learning and think-aloud protocols. In Blikstein, P., Van Aalst, J., Kizito, R., & Brennan, K. (Eds.), Proceedings of the 17th International Conference of the Learning Sciences - ICLS 2023 (pp. 982-985). International Society of the Learning Sciences.
Abstract: In this study, we used machine learning models to detect the goal setting and planning activities in self-regulated learning (SRL) based on the linguistic features of think-aloud transcripts. Specifically, we trained six types of machine learning models (i.e., decision tree, Gradient boosted decision tree, random forest, logistic regression, support vector machine, and neural network) on 2,792 think-aloud segments of medical students, who were asked to think out loud as they diagnosed virtual patients in a computer-simulated environment. The results suggested that machine learning models, especially Gradient boosted decision tree and neural network, could make accurate predictions. This study shows the possibility of using machine learning to free researchers from the labor-intensive work of coding think-aloud transcripts. This study also informs practitioners about automatically detecting students’ SRL activities in real-time as they think aloud in learning, making the provision of timely feedback possible.
Description: Short Paper
URI: https://doi.org/10.22318/icls2023.101292
https://repository.isls.org//handle/1/10375
Appears in Collections:ISLS Annual Meeting 2023

Files in This Item:
File SizeFormat 
ICLS2023_982-985.pdf641.61 kBAdobe PDFView/Open


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