.png){kind=icon}
.png){kind=icon}
Return
Enabling data-driven peer help for extracurricular learning: system design and initial implementation in junior mathematics
.png){kind=icon}
PRE
AI
DOI：10.1080/10494820.2025.2536576.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rapid advancement of educational technologies has reshaped informal and extracurricular learning, particularly in regions like Hokkaido, Japan, where demographic shifts have increased reliance on remote learning. While these settings offer flexibility and accessibility, they often lack structured opportunities for interpersonal interactions, posing challenges to social interactional learning and timely assistance. This study proposes an adaptive peer learning system that fosters personalized and collaborative learning beyond traditional classrooms. The system employs a knowledge graph-based recommendation algorithm aligning with dynamically updated learners’ knowledge profiles, thereby addressing the limitations of static recommendation methods and biases in peer selection. A two-month exploratory study in a junior high mathematics course investigated the system’s effectiveness. Results demonstrated its ability to expand and stabilize peer-learning participation and its potential to enhance learning performance. User feedback highlighted its capacity to promote learner autonomy, equal participation, and flexible learning opportunities in extracurricular contexts. This study underscores how dynamic feedback loops with iterative data accumulation, combined with interpretable features such as knowledge visualization, can support an adaptive, inclusive, and socially enriched learning environment.
Keywords:
Extracurricular learning
peer learning
peer help
data-driven system
secondary education
mathematics learning
Journal
IF:
5.3
Papers:
2.7K
Citations:
9.0K