Development of Tutoring Assistance Framework Using Machine Learning Technology for Teachers

Abstract

This paper proposes a framework for tutoring assistance to tackle increasing student dropout rates. Student dropouts in higher education institutions, such as universities, often result in an increase in tutors’ workload. Currently, educational assistance is focused on supporting the students’ learning, and the main purpose of this assistance is an acceleration of the learning process. Although student assistance is undoubtedly of great importance, offering assistance to teachers who also have a tutoring role is equally important. The purpose of our framework for assistance is to detect students at risk for dropout, after which an alert is sent to the tutors. The alert encourages tutors to take timely action to avoid student dropouts. This paper describes the enhanced framework implementation, its experimental use, and the results.