Research on Collaborative Interaction Mode of Intelligent Physical Education Teaching Based on Cloud Platform under the View of Digitization

Abstract

The cloud computing platform’s data and information sharing, computational efficiency, and service convenience provide strong support for collaborative learning and teaching and promote the high-level development of education digitalization. This paper designs a smart sports teaching system based on the collaborative cloud computing-assisted teaching platform, enabling a three-dimensional collaborative smart sports teaching mode. This paper focuses on the intelligent recommendation problem of sports learning resources and optimizes the teaching system by constructing a collaborative filtering recommendation model based on a graph convolutional neural network. The AUC, MRR, NDGG@1, and NDGG@2 index values and loss values of this paper’s sports learning resources recommendation model are 0.789, 0.904, 0.797, 0.934, and 0.54, respectively, which are better than PinSage, CASER, DIN, and MCR models. The model in this paper has shown that it can work by improving NDCG@2 and AUC values in a variety of sparsity learner groups in a way that is both effective and stable. The T-test P-values of the post-experimental badminton skill levels of the experimental group adopting the cloud-based 3D collaborative intelligent physical education teaching model and the control group adopting the traditional teaching model are all less than 0.05, which is a significant difference. This indicates that the physical education teaching mode in this paper is better than the traditional teaching mode and more conducive to improving students’ sports technology.

Keywords

• collaborative filtering recommendation algorithm
• graph convolutional neural network
• cloud platform
• intelligent sports teaching
• 97p10