The Application of Emotion Detection System (EmoD) in Online Learning

Abstract

Measuring the student's emotions can help the teacher increase student engagement during Teaching and Learning (TnL) and simultaneously help the teachers measure the effectiveness of the TnL activity. However, most online TnL activities do not consider the learner's emotions and affection. Not all facial recognition methods can detect emotion, especially in real-time. To overcome this problem, we developed an Emotion Detection System (EmoD) to recognize and identify the emotion based on its class using ResNet 50. Using the System Usability Scale (SUS), the EmoD was reported as acceptable (mean = 78.5). Other than that, the EmoD system can be utilized for online consultation and classes. In the future, the EmoD will be enhanced so that it can be used to detect more than one user to make it more useful for online learning.