Analyzing Student Performance and Interaction in Online Learning Environments using LMS Log Data

The COVID-19 pandemic, online learning has become more common, and learning management systems (LMS) are increasingly used. Students' study habits and patterns can be analyzed with the help of LMS log data, which can be used to identify at-risk students and motivate high-performing ones with activities. By analyzing the generated LMS log files, it creates models for the early prediction of students' performance in solving LMS assignments. Instead of predicting the exact student's mark using LMS assignments, this study uses datasets created with all online classes for a single academic year to identify at-risk, fail, and excellent students in the early stages of the online class. The study shows that decision tree, logistic regression, naive Bayes, multilayer perceptron neural network, and support vector machine models can be used to make accurate prediction models for students who are at risk, who fail, and who do well in the online class at different times. The MLP model has the best results, with an accuracy of 90.1 % after half of the online class has been completed.