Artificial intelligence-based group decision making to improve knowledge transfer: the case of distance learning in higher education

Abstract

ABSTRACT In this paper, we propose a method based on multicriteria classification and a dominancebased rough set approach (DRSA) to support teachers in decision making. The objective is to use teachers’ knowledge and preferences to identify ‘atrisk students’, i.e. students who are likely to drop out, and ‘leader students’, i.e. students who are likely to help their peers, in distance learning. The proposed method is composed of two phases: phase I builds collective decision rules from teachers’ preferences, and phase II classifies students into two decision classes: ‘atrisk students’ and ‘leader students’. This method was designed, tested, and validated in higher education, with teachers who have acquired rich experience in teaching in online-synchronous mode since the Covid-19 pandemic.