# Detecting Behavioral Engagement of Students in the Wild Based on Contextual and Visual Data

## TL;DR

This study presents a two-phase multi-modal approach combining contextual logs and visual data to accurately detect students' behavioral engagement in real-world educational settings, demonstrating improved performance and cross-context applicability.

## Contribution

It introduces a novel two-phase framework that integrates contextual and visual data for behavioral engagement detection, showing enhanced accuracy and generalizability.

## Key findings

- Increased F1-score from 0.77 to 0.82 with the proposed method.
- Effective cross-classroom and cross-platform applicability.
- Demonstrated robustness across different student groups and subjects.

## Abstract

To investigate the detection of students' behavioral engagement (On-Task vs. Off-Task), we propose a two-phase approach in this study. In Phase 1, contextual logs (URLs) are utilized to assess active usage of the content platform. If there is active use, the appearance information is utilized in Phase 2 to infer behavioral engagement. Incorporating the contextual information improved the overall F1-scores from 0.77 to 0.82. Our cross-classroom and cross-platform experiments showed the proposed generic and multi-modal behavioral engagement models' applicability to a different set of students or different subject areas.