Deep learning for student engagement analysis in educational psychology

Area of Science:

• Educational Psychology
• Artificial Intelligence
• Machine Learning

Background:

• Student engagement is crucial for academic success.
• Traditional models fail to capture the dynamic nature of engagement.
• There is a need for advanced analytical tools in educational psychology.

Purpose of the Study:

• To present an innovative deep learning framework, the Engagement Dynamics Forecaster.
• To analyze and predict student engagement patterns with enhanced accuracy and depth.
• To offer a comprehensive and adaptive approach to understanding student engagement.

Main Methods:

• The framework integrates three components: Manifold Constrained Interaction Filter, Agent Driven Sequential Planner, and Uncertainty Propagation Regularizer.
• It employs constrained optimization refinement and agent-based decision scheduling.
• Combines domain-specific insights with deep learning methodologies.

Main Results:

• Empirical results demonstrate the model's efficacy in linking engagement theory with practice.
• The framework accurately forecasts engagement dynamics.
• Provides valuable tools for educators and researchers.

Conclusions:

• The Engagement Dynamics Forecaster offers a novel approach to student engagement analysis.
• It holds significant promise for advancing educational strategies and interventions.
• Contributes to improved educational outcomes through better understanding of engagement.