Using machine learning to predict students' academic performance during Covid-19

Abstract

COVID-19 pandemic has affected various sectors of the global economy including the abrupt closure of schools in March 2020 in Ghana. This sudden closure has led to a revamp in online teaching and learning across most institutions with learners submitting their assignments and taking their assessments on various learning management systems while at home. In this study, we used classification algorithms to investigate features and predict the academic performance of students during the pandemic. We collected data from students in the Department of ICT Education of the University of Education, Winneba during the COVID-19 period using carefully selected attributes that could affect their exams score. The results detailed dominant attributes that affected students' performance with Random Forest, Random Tree, Na�ve Bayes and J48 Decision Tree algorithms further analysed for accuracy, confusion matrix and the ROC Curve. After detailed analysis, we observed that the accuracy of a classifier alone is not indicative enough of its performance. � 2021 IEEE