Supporting Instructors with Course Attendance and Quality Prediction in Synchronous Learning

The massive adoption of artificial intelligence has opened up the opportunity for a range of intelligent technologies that can support education. Empowering instructors with tools able to early predict the attendance and quality of their courses and consequently make prompt adjustments is one of them. However, the potential of these tools has by no means been researched exhaustively within synchronous courses, with prior work mostly focused on courses based on pre-recorded lectures. In this paper, we examine the predictive power of machine-learning models on the future attendance of students to synchronous courses and their perceived quality. To this end, we leverage both (i) attendance records to online real-time lectures within courses of a large public university, and (ii) responses to course quality questionnaires completed at the end of each course. Experiments show that our models can accurately and early predict attendance to courses and key aspects characterizing course quality (e.g., study workload). Our findings confirm the potential of intelligent models to support instructors in managing and promptly reacting within their courses, to increase engagement and reduce dropout.