Prediction of Student Success using Student Engagement with Learning Management System

There has been a surge in student failure rates in blended-learning courses in recent times, which has generated considerable research interests. Engagement is identified as one of the core metrics for measuring students’ success or failure in any learning system. This study utilizes machine learning algorithms on students’ log-file data collected from an LMS to predict student success and increase the students’ throughput rates. The machine learning predictive models considered in this study are the Naive Bayes Classifier, Decision Tree, Gradient Boosting Tree, Linear Logistic Regression, Random Forest, Multilayer Perceptron Neural Network, and Support Vector Machines. The results provide an automatic predictive model for early detection of students at risk of failing for timely instructor intervention. The result serves as a feedback tool on learning for an increase in student performance. The machine learning algorithms’ performances were evaluated using accuracy, precision, recall, and ROC-AUC for the best performing predictive model.