Prediction of student engagement using deep learning-based student face expression detection method (DL-SFEDM)

Nowadays, Academics and teachers have paid a lot of attention to computational thinking (CT) because of the wonderful opportunities it presents for developing students' problem-solving abilities, which are in high demand in a technologically advanced world. However, research has shown that educators lack a solid grasp of CT and often misunderstand its idea, which might impede their implementation of these initiatives. This problem is even worse because very little research has explored ways to engage students with their learning. In this paper, a Deep Learning-Based Student Face Expression Detection Method (DL-SFEDM) is a method for finding out how engaged students are with online lecture videos that don't depend on data generated by educational management systems. To conduct the classroom engagement analysis, the whole class is treated as a single group and their corresponding group engagement score is calculated. Emotions play a crucial part in learning. Computer vision (CV)-based approaches evaluate online and offline lecture recordings and extract students' emotions. This expressive emotion analysis looks at how Students feel in four distinct states: positive, negative, neutral, and negative. The experimental results show that the proposed DL-SFEDM enhances the teaching technique, motivates better student learning and increases student computational thinking.