Developing an early-warning system through robotic process automation: Are intelligent tutoring robots as effective as human teachers?

Abstract

ABSTRACT Artificial intelligence aims to restructure and process re-engineering education and teaching processes and accelerate the evolution of the whole education system from information to intelligence. Robotic Process Automation (RPA) robots learn by observing people at work, analyzing user processes repeatedly, and adjusting or correcting automated processes. By using Natural Language Processing (NLP) and Machine Learning (ML), knowledge representation, inference, large-scale parallel computing, and Rapid Domain Adaptation, RPA robots can automatically extract the data needed for decision-making and continuously learn from users’ feedback. We have used RPA and predictive analytics to provide distance learning students with the Intelligent Tutoring Robot (ITR), which can provide an automatic response. By optimizing the ITR in the above context, we have examined the feasibility of transforming a prediction model, using a student learning database, into an early-warning system. This article adopts the randomized control-group pretest-posttest design, dividing 123 students into a control group to describe interactions between ITR and students and experimental groups to describe interactions between human teachers and students. The findings present no significant difference between the control and the experimental groups in terms of academic performance, however higher average marks were achieved in the former group.