Online Student Learning Experience

These days, the rise of online education offers unprecedented access to education and learning experiences that copy learning in the classroom. But, student involvement, assessment of learning experiences, and feedback to students have become complex as part of such significant growth. Conventional ways used to measure student attention and affect have become ineffective as educational activities have migrated online requiring new approaches to improve online learning quality. This paper presents a new model for evaluating student engagement and learning experience in online education using computer vision and deep learning. The system assesses student engagement and feelings during lectures through the analysis of real-time facial expressions. To ensure real-time feedback, only comments from attentive students are analysed for sentiment using the VADER tool. The project is planned with sufficient time and towards engagement detector. So, we are focused on enhancement of robustness of engagement detection by using visual analysis of facial expressions in combination with other instruments such as eyetracking technology, physiological sensors (e.g. heart rate variability, skin conductance), and Natural Language Processing (NLP) techniques. The researchers examine machine learning models trained on a large dataset of student interactions and performance data. By leveraging these approaches, teachers will have the data to understand how engaged students and their emotions are during class. This will allow them to improve their way of teaching and online learning in general.