Deep learning based multi attribute evaluation for holistic student assessment in physical education

Area of Science:

• Educational Technology
• Artificial Intelligence in Education
• Sports Science Assessment

Background:

• Traditional physical education assessments are often one-dimensional and lack comprehensiveness.
• Existing methods struggle to integrate diverse student performance data effectively.
• There is a need for more objective, scalable, and personalized evaluation tools in physical education.

Purpose of the Study:

• To propose a deep learning-based multi-attribute user evaluation model for holistic student assessment in physical education.
• To address the limitations of traditional, one-dimensional assessment approaches.
• To enhance the accuracy, flexibility, and objectivity of student performance evaluation.

Main Methods:

• A ten-step methodology involving data collection, preparation, model construction, and deployment.
• Utilizing deep learning for multi-attribute user evaluation modelling.
• Integrating diverse data: physical activities, cognitive tasks, emotional responses, and social interactions.

Main Results:

• The developed model demonstrates high efficacy with improved accuracy and reduced errors.
• Experimental investigation confirms the model's robustness, indicated by a low mean score.
• The model provides personalized feedback, enhancing decision-making and educational outcomes.

Conclusions:

• The deep learning model offers automated, objective, and scalable attributes for superior student assessment in physical education.
• Visualization tools like heatmaps aid in performance monitoring and adaptive adjustments.
• This approach effectively tackles multifaceted challenges, enabling targeted interventions for student advancement.