Student Education Analysis of e-Learning during COVID-19 Using Support Regression Random Forest Algorithm

COVID-19 pandemic has had a profound effect on all part of our existence, including schooling. Due to COVID-19, there was a noticeable increase in online education and the remote use of technological devices as instructional approaches. Most colleges and universities offered online courses in an effort to lessen the pandemic’s impact on the field of education. However, it was crucial to make sure that learners remain appropriately involved throughout online lectures in order to make these digital learning sessions dynamic and equivalent to the conventional physical classrooms. In this work, a novel approach has been introduced based on machine learning that tracks the mental states of learners in instantaneous fashion, and classifies it into six categories including joyful, disappointed, astonished, dissatisfied, anxious, and neutral. This is accomplished by the combination of algorithms such as Support Vector Regression, XGBoost and Random Forest to examine the data from facial reference identification and expression recognition. The suggested computerised technique is guaranteed to help colleges and universities in creating a better and more creative online instruction strategy.