GOAT: a novel global-local optimized graph transformer framework for predicting student performance in collaborative learning DOI Creative Commons
Tianhao Peng,

Qiang Yue,

Yu Liang

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 21, 2025

Collaborative learning is a prevalent method, and modeling predicting student performance in such paradigms an important task. Most current methods analyze this complex task solely based on the frequency of activities, overlooking rich spatial temporal features present these as well diverse textual content provided by various artifacts. To address challenges, we choose software engineering course study subject, where students are required to team up complete project together. In paper, propose novel Global-local Optimized grAph Transformer framework for collaborative learning, termed GOAT. Specifically, first construct dynamic knowledge concept-enhanced interaction graphs with nodes representing both relevant concepts, edges illustrating interactions. Additionally, incorporate spatial-aware temporal-aware modules capture respective information, enabling interactions within across teams over time. A global-local optimization module introduced model intricate relationships between teams, highlighting commonalities differences among members. Our backed theoretical analysis validated through extensive experiments real-world datasets, which demonstrate its superiority existing methods.

Язык: Английский

The Future of Education: A Multi-Layered Metaverse Classroom Model for Immersive and Inclusive Learning DOI Creative Commons
Leyli Nouraei Yeganeh, Nicole S. Fenty, Yu Chen

и другие.

Future Internet, Год журнала: 2025, Номер 17(2), С. 63 - 63

Опубликована: Фев. 4, 2025

Modern education faces persistent challenges, including disengagement, inequitable access to learning resources, and the lack of personalized instruction, particularly in virtual environments. In this perspective, we envision a transformative Metaverse classroom model, Multi-layered Immersive Learning Environment (Meta-MILE) address these critical issues. The Meta-MILE framework integrates essential components such as immersive infrastructure, interactions, social collaboration, advanced assessment techniques enhance student engagement inclusivity. By leveraging three-dimensional (3D) environments, artificial intelligence (AI)-driven personalization, gamified pathways, scenario-based evaluations, model offers tailored experiences that traditional classrooms often struggle achieve. Acknowledging potential challenges accessibility, infrastructure demands, data security, study proposed practical strategies ensure equitable safe interactions within Metaverse. Empirical findings from our pilot experiment demonstrated framework’s effectiveness improving skill acquisition, with broader implications for educational policy competency-based, experiential approaches. Looking ahead, advocate ongoing research validate long-term outcomes technological advancements make more accessible secure. Our perspective underscores shaping inclusive, future-ready environments capable meeting diverse needs learners worldwide.

Язык: Английский

Процитировано

3

GOAT: a novel global-local optimized graph transformer framework for predicting student performance in collaborative learning DOI Creative Commons
Tianhao Peng,

Qiang Yue,

Yu Liang

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 21, 2025

Collaborative learning is a prevalent method, and modeling predicting student performance in such paradigms an important task. Most current methods analyze this complex task solely based on the frequency of activities, overlooking rich spatial temporal features present these as well diverse textual content provided by various artifacts. To address challenges, we choose software engineering course study subject, where students are required to team up complete project together. In paper, propose novel Global-local Optimized grAph Transformer framework for collaborative learning, termed GOAT. Specifically, first construct dynamic knowledge concept-enhanced interaction graphs with nodes representing both relevant concepts, edges illustrating interactions. Additionally, incorporate spatial-aware temporal-aware modules capture respective information, enabling interactions within across teams over time. A global-local optimization module introduced model intricate relationships between teams, highlighting commonalities differences among members. Our backed theoretical analysis validated through extensive experiments real-world datasets, which demonstrate its superiority existing methods.

Язык: Английский

Процитировано

0