Development and Evaluation of a Real-Time Emotion Detection System to Enhance Student Interaction

Gerlan Apriandy Manu*, State University of Malang
Punaji Setyosari, State University of Malang
Saida Ulfa, State University of Malang
Henry Praherdhiono, State University of Malang

https://doi.org/10.9743/JEO.2026.23.1.9

Abstract

This research explores the development of a real-time emotion detection system to improve engagement in online learning. The system uses Convolutional Neural Networks (CNN) to identify five emotions: happy, sad, angry, surprised, and neutral via webcam during virtual classes. Tested with 30 students in an Artificial Intelligence course, it achieved 86.4% accuracy, excelling in detecting happy and neutral states. Instructors used emotional feedback to adapt teaching dynamically, enhancing learning experiences and satisfaction. Feedback showed that 88% of students felt more motivated and engaged. This study highlights the potential of emotion-based tools in bridging gaps between online and traditional education.

Keywords: emotion detection deep learning, Convolutional Neural Network (CNN), student engagement, adaptive learning, Artificial Intelligence in education.

*Corresponding author : gerlan.apriandy.2301219@students.um.ac.id