Personalized digital media teaching system based on AJS-TRes-GRU: an ai-driven path integrating theory and practice
Data and Metadata · 16.04.2026 · 5 · 813
DOI: 10.56294/dm2026813 ↗Kurz gefasst
Fachartikel in der Zeitschrift Data and Metadata, online erschienen am 16.04.2026. Verfasst von Xiaolu Fan und 2 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Digital media education increasingly requires the integration of theoretical knowledge with practical application to enhance creativity and technical skills. Conventional teaching methods often fail to provide personalized guidance or maintain style consistency, limiting learner engagement and skill development.
Objective
To address these challenges, an Artificial Intelligence (AI) driven system was developed to bridge theory and practice, improving learner engagement, creative efficiency, and skill acquisition.
Method
The system employs a Novel Artificial Jellyfish Search Optimizer-driven Temporal Residual Gated Recurrent Unit (AJS-TRes-GRU) to optimize automated content generation, ensuring style consistency and adaptive instruction. Input data, including student engagement metrics and digital media content from educational platforms and social media, are preprocessed using image resizing and median filtering. The WikiArt Dataset was utilized for training the system, providing a diverse collection of high-quality artwork to ensure the style-consistent generation of digital media. The TRes-GRU framework automates style-consistent digital media generation, fine-tunes GRU parameters via AJSO, and builds detailed learner profiles. Integration with Creative Intelligence Cloud (CIC) supports hands-on production with personalized guidance.
Result
The model achieved an Inception Score (IS) of 5,2, SSIM of 0,9, and FID of 9,17, demonstrating superior performance over prior approaches using python simulation tool; paired t-test results showed narrow 95 % confidence intervals, confirming the model’s reliability and consistency. Overall, the system produces realistic, structurally accurate, and style-consistent digital media, significantly improving content quality, learner engagement, and skill acquisition.
Conclusion
By combining AI-driven content generation with adaptive learning frameworks, the research creates an immersive, personalized digital media education environment, effectively preparing learners for evolving industry demands.
Zitieren
APA
Fan, X., Han, W., & Wang, D. (2026). Personalized digital media teaching system based on AJS-TRes-GRU: an ai-driven path integrating theory and practice. Data and Metadata, 5, 813. https://doi.org/10.56294/dm2026813
Vancouver
Fan X, Han W, Wang D. Personalized digital media teaching system based on AJS-TRes-GRU: an ai-driven path integrating theory and practice. Data and Metadata. 2026;5:813. doi:10.56294/dm2026813
BibTeX
@article{Fan2026,
author = {Fan, Xiaolu and Han, Weinan and Wang, Dongzhen},
title = {Personalized digital media teaching system based on AJS-TRes-GRU: an ai-driven path integrating theory and practice},
journal = {Data and Metadata},
year = {2026},
volume = {5},
pages = {813},
doi = {10.56294/dm2026813}
}
EUVIMED