Automated Student Emotion Analysis During Online Classes Using Convolutional Neural Network

Abstract

In non-verbal communication, facial emotions play a very crucial role. Facial recognition can be useful in various ways, such as understanding people better and using the collected data in various fields. In an e-learning platform, students’ facial expressions determine their comprehension levels. Students’ facial emotions can have a favorable or unfavorable impact on their academic performance. As a result, instructors need to create a positive, emotionally secure classroom environment to optimize student learning. In this paper, a novel Facial Emotion Recognition for improving our understanding of students during e-learning is proposed. Suggested model detects different students’ facial emotions such as anger, disgust, fear, happiness, sadness, surprise, and neutral and utilizing them for better teaching and learning during a lecture in an e-learning platform. Convolutional neural networks (CNNs) have been used for detecting facial emotions of students in e-learning platforms, and the proposed model shows an outcome of test accuracy of 67.5%.