Title: Visual learning quality evaluation of online teaching platform based on MOOC learning behaviour big data

Authors: Zhi Zhang

Addresses: College of Preschool Education, Chongqing Aerospace Polytechnic, Chongqing, Jiangjin District, China

Abstract: To enhance the precision and efficacy of assessing visual learning quality on online educational platforms, a methodology leveraging MOOC-derived big data on learning behaviours has been developed. Initially, a text feature vector that accounts for word significance is constructed through the computation of word frequency and inverse document frequency. Subsequently, by integrating the vectorised representation of the DOM tree structure, predictions are made regarding the attribute tags associated with learners' behavioural categories on the platform. Lastly, the significance of these behavioural category attribute tags is scrutinised, and a logical regression model, incorporating the sigmoid function and regularisation techniques, is employed to conduct the visual learning quality assessment. Experimental outcomes reveal that the proposed approach boasts a prediction accuracy exceeding 90%, with a peak average processing time of merely 19.6 seconds, demonstrating its effective applicability.

Keywords: MOOC learning behaviour; online teaching platform; visual learning; quality evaluation.

DOI: 10.1504/IJCAT.2024.146137

International Journal of Computer Applications in Technology, 2024 Vol.75 No.2/3/4, pp.122 - 129

Received: 23 Aug 2024
Accepted: 02 Jan 2025
Published online: 07 May 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article