Title: A secure and reliable large-scale online physical education solution based on deep learning for cloud computing

Authors: Wenbo Song

Addresses: Jilin Normal University, Siping, Jilin, China

Abstract: With the continuous expansion of online education scale, its existing problems have gradually emerged. Therefore, it is necessary to develop a secure and reliable large-scale online physical education solution, cultivate higher-order thinking and guide students to learn effectively. This article studies cloud computing and deep learning models. And it proposes an improved model based on Graph Knowledge Tracing (GKT). The improved model can better discover causal relationships and use network structure to track students' knowledge status. Finally, experiments are conducted on the improved model and the classical model, proving that the improved model based on GKT performs well. The improved model based on GKT outperforms other models and has superior convergence speed and accuracy. It has good practicality for large-scale online physical education, helps to obtain interactive information of implicit knowledge in the scheme, and increases the richness of education.

Keywords: online physical education; cloud computing; deep learning; GKT.

DOI: 10.1504/IJCAT.2025.149359

International Journal of Computer Applications in Technology, 2025 Vol.76 No.3/4, pp.155 - 165

Received: 12 Apr 2024
Accepted: 03 Oct 2024
Published online: 27 Oct 2025 *

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