Date
Publisher
arXiv
Analyzing student behavior in educational scenarios is crucial for enhancing teaching quality and student engagement. Existing AI-based models often rely on classroom video footage to identify and analyze student behavior. While these video-based methods can partially capture and analyze student actions, they struggle to accurately track each student's actions in physical education classes, which take place in outdoor, open spaces with diverse activities, and are challenging to generalize to the specialized technical movements involved in these settings. Furthermore, current methods typically lack the ability to integrate specialized pedagogical knowledge, limiting their ability to provide in-depth insights into student behavior and offer feedback for optimizing instructional design. To address these limitations, we propose a unified end-to-end framework that leverages human activity recognition technologies based on motion signals, combined with advanced large language models, to conduct more detailed analyses and feedback of student behavior in physical education classes. Our framework begins with the teacher's instructional designs and the motion signals from students during physical education sessions, ultimately generating automated reports with teaching insights and suggestions for improving both learning and class instructions. This solution provides a motion signal-based approach for analyzing student behavior and optimizing instructional design tailored to physical education classes. Experimental results demonstrate that our framework can accurately identify student behaviors and produce meaningful pedagogical insights.
Study at a glance
- Question: Can wearable motion sensors and large language models accurately analyze student physical education behaviors and generate helpful teaching feedback?
- Setting and data: 72 university student volunteers (freshmen and sophomores) in basketball physical education classes at a university in China; public activity recognition datasets.
- Method: System development, benchmark evaluation, and a real classroom intervention comparing student metrics before and after feedback-driven regrouping.
Why it matters
Traditional physical education observation is labor-intensive, while video-based AI struggles to track complex movements across open spaces and lacks pedagogical knowledge to provide meaningful teaching feedback.
Key findings
- In a real classroom experiment, model-guided student regrouping increased core action proportion by 4.40 percentage points and core action duration by 5.46 percentage points.
- The framework's activity recognition model achieved 92.30% accuracy on a benchmark dataset and reached 79.69% accuracy on real physical education movements with few-shot fine-tuning.
- The action quality assessment model achieved a 0.63 mean squared error and 0.51 correlation with human expert ratings, outperforming a video-based baseline model.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Why use AI?
