Assessing learning engagement based on facial expression recognition in MOOC's scenario

Area of Science:

• Educational Technology
• Computer Science
• Artificial Intelligence

Background:

• Online learning is a prevalent educational modality requiring continuous learner motivation and engagement.
• Assessing learner engagement in real-time is challenging for supervisors in online environments.
• Distance learners often require self-motivation, highlighting the need for effective engagement monitoring tools.

Purpose of the Study:

• To propose a novel framework for real-time assessment of online learning engagement.
• To introduce an intelligent video surveillance technique incorporating facial expression recognition.
• To develop a domain adaptation-based facial expression recognition method suitable for MOOCs.

Main Methods:

• Implementation of a framework for learning engagement assessment using intelligent video surveillance.
• Integration of facial expression recognition to capture learners' real-time emotional changes.
• Development of a new facial expression recognition algorithm employing domain adaptation for MOOC contexts.

Main Results:

• The proposed framework demonstrated effectiveness in assessing learners' engagement in online learning settings.
• Experimental results validated the system's capability to monitor emotional shifts indicative of engagement.
• Comparisons with existing state-of-the-art methods confirmed the superiority of the proposed facial emotion recognition technique.

Conclusions:

• The developed framework offers a viable solution for real-time online learning engagement assessment.
• Facial expression recognition, particularly with domain adaptation, is a promising approach for monitoring learner affect in MOOCs.
• This technology can aid supervisors in understanding and potentially improving the online learning experience.