Deep Learning for Enhanced Education Quality: Assessing Student Engagement and Emotional States

This paper presents a comprehensive study on the application of deep learning techniques in education to enhance instructional quality. While deep learning has been extensively explored in various domains, its potential for reporting and improving instruction in education remains largely untapped. Student engagement and interest are crucial factors in the learning process and future success. However, students often face challenges in maintaining active participation, hindering their learning experience and overall teaching effectiveness. Current research on utilizing deep learning for assessing student engagement in education is limited. To bridge this gap, this study introduces an innovative approach that employs deep learning models to monitor students’ emotional states and interest levels in classes, providing valuable insights to instructors. The emotion detection model achieved an impressive accuracy of 92.16%, enabling accurate assessment of students’ emotional responses during educational activities. Moreover, the face recognition model achieved a remarkable accuracy of 98.10%, facilitating reliable identification of students and tracking their presence and engagement. By leveraging these deep learning models, this study aims to establish a student-centered learning environment that enhances education quality. The findings contribute to effective student support and improvement of educational processes by harnessing the power of deep learning techniques in the field of education.