EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Personalized physical education teaching content generation based on generative AI

Peng Zhan

Data and Metadata · 12.03.2026 · 5 · 797

DOI: 10.56294/dm2026797 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Data and Metadata, online erschienen am 12.03.2026. Verfasst von Peng Zhan. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Physical Education (Phy-Ed) plays a crucial role in fostering holistic student development, including physical fitness, motor skills, and cognitive engagement. However, conventional teaching approaches lack personalization and fail to dynamically adapt to individual learner characteristics, thereby limiting instructional effectiveness.

Objective

This research aims to develop an intelligent generative framework for personalized Phy-Ed content generation using multimodal learner data and advanced optimization techniques.

Method

The proposed Phy-Ed-Generative AI framework integrates a Transformer-based Generative Adversarial Network (TGAN) optimized using the Mantis Search Algorithm (MSA). The Phy-Ed-Multimodal dataset, comprising 300,000 records with 28 features, includes demographic attributes, fitness metrics, and movement patterns. Data preprocessing involves Min-Max normalization and outlier detection, followed by Principal Component Analysis (PCA) for feature extraction and dimensionality reduction. The Transformer module captures sequential dependencies in learner performance, while the GAN generates realistic, context-specific instructional modules. MSA optimizes TGAN hyperparameters to enhance generation efficiency and accuracy.

Results

Experimental evaluation demonstrates that the proposed MSA-TGAN framework achieves an overall accuracy of 95,2%, outperforming conventional GAN and baseline machine learning models. The system significantly improves motor skill performance, physical fitness progression, participation rate, student satisfaction, and overall teaching effectiveness.

Conclusions

The proposed framework provides a scalable and adaptive solution for personalized Phy-Ed instruction. By integrating multimodal learning data with generative AI and optimization strategies, the model effectively bridges personalization gaps in traditional Phy-Ed methodologies. .

Zitieren

APA

Zhan, P. (2026). Personalized physical education teaching content generation based on generative AI. Data and Metadata, 5, 797. https://doi.org/10.56294/dm2026797

Vancouver

Zhan P. Personalized physical education teaching content generation based on generative AI. Data and Metadata. 2026;5:797. doi:10.56294/dm2026797

BibTeX
@article{Zhan2026,
  author = {Zhan, Peng},
  title = {Personalized physical education teaching content generation based on generative AI},
  journal = {Data and Metadata},
  year = {2026},
  volume = {5},
  pages = {797},
  doi = {10.56294/dm2026797}
}
Kostenlos starten →