A data mining approach to classifying e-learning satisfaction of higher education students: a Philippine case

Abstract

E-learning has become increasingly important for higher education institutions. It offers an alternative mode of learning for educational institutions during critical situations such as the COVID-19 pandemic. While e-learning has gained growing attention in the current literature, a significant gap is left unaddressed for emerging economies, particularly the Philippines. In this paper, the factors of e-learning in a higher education institution in the Philippines are analysed. A data mining approach is used to predict the satisfaction of higher education students given eleven features of the subjects. Four classifiers: 1) logistic regression; 2) support vector machine; 3) multilayer perceptron; 4) decision tree, are used to develop the predictive models. The findings reveal that the features considered in this paper can be used to accurately predict the student satisfaction towards e-learning of higher education students in the Philippines.