Predicting Student Performance from Video-Based Learning System: A Case Study

The major impact of the COVID-19 pandemic on the shift of education norms from physical classroom learning to MOOCs (Massive Open Online Courses) could accelerate the big data era growth for the e-learning platform. This circumstance has provided an opportunity for a teacher to use MOOC data to help students learn and perform better. Moreover, this research study goal is to propose a combination of machine learning algorithms and the feature selection benefit with the SMOTE (Synthetic Minority Oversampling Technique) algorithm for balancing the output features number to predict student performance in a video-based learning platform. As a result, the proposed machine learning classifier, Naïve Bayes algorithm with the combination of chi-square test and SMOTE has shown the highest accuracy in prediction of more than 90%. Results by the proposed classifier with feature selection and SMOTE have outperformed the traditional machine learning classifiers