Identifying student group profiles for diagnostic feedback using snap-drift modal learning neural network

Abstract

The aim of this paper is to propose a novel method for identifying student group profiles based on student responses to a set of multiple choice questions for the purpose of constructing diagnostic feedback using snap-drift modal learning neural network. The proposed method is capable of supporting tutors without the knowledge of machine learning in identifying useful student groups and constructing diagnostic feedback. Trials were conducted and analysis of the result showed that the  snap-drift modal learning neural network was able to identify distinct student groups and represented student group profiles were helpful in revealing gaps of understanding and misconceptions that facilitate construction of diagnostic feedback. Moreover, the result showed that all student responses gathered were assigned to their appropriate student group profiles and the diagnostic feedback constructed based on the identified student group profiles had a positive impact on improving the learning performance of the students.