Predicting Student Performance Through Classroom Concentration: A Review on Machine Learning and Computer Vision Approaches

Student performance prediction is a complex problem in which a computer predicts student’s future performance while they engage in their studies. To enable effective educational interventions during a course, accurate early forecasts of a student’s performance is essential. The review proposed a silent method and the latest data for capturing student behaviour is classroom video monitoring. For student performance prediction dynamics to be understood and subsequently improved, it is essential to understand students’ attention spans and what kinds of behaviours may advise the absence of attention. The ability to detect student attention is one of the features of computer vision systems used to monitor classrooms. This review included the pertinent Educational Data Mining literature on identifying dropouts and students at risk from classroom data. For example, during class hours students are to concentrate, focus and give all their attention in class. A machine learning approach can be used to rate each student’s concentration with the help of computer vision algorithms since most studies use student data from colleges such as quiz marks and exam scores. The assessment outcomes revealed that a variety of Machine Learning (ML) and computer vision techniques are developed to understand and address the main problems, including guessing about to-fail learners and student dropouts. It has been validated that machine learning systems are important for detecting at-risk students and dropout frequencies which help for improving student success. To find the basics that can have an impact on how well students perform academically machine learning algorithms can be used to study a student’s daily routine such as their study habits, social interactions, and extracurricular activities. Since it applies the most relevant advanced methods for each specific task, this student concentration score using computer vision offers the ideal answer for monitoring classrooms. The integration of facial recognition algorithms and machine learning provides an assuring approach for monitoring classroom engagement and enhancing educational interventions.