A Comparative Study of Classification Algorithms of Moodle Course Logfile using Weka Tool

Learning Management Systems (LMSs) have been widely
used in the deployment of e-learning in higher education
institutions. One of the most famous LMS used is Moodle. In
Moodle environment, classification has been used for several
reasons, including finding students who share similar traits and
forecasting student performance. Therefore, this study looks at
two classification algorithms that were used on a dataset
gathered from a Moodle LMS course logfile. The goal is to
conduct a thorough theoretical and experimental examination of
classification data mining techniques, as well as a comparison
study, to determine which methodology is the best for
identifying student performance with the support of their
engagement, behavior, and personality during different
activities of the course. The algorithms under investigation are
Naive Bayes (NB) and Random Forest (RF). The performance
of the classification of the two algorithms is compared using the
tool Weka “Waikato Environment for Knowledge Analysis” as
an open-source software package that includes data preparation,
algorithm implementation and visualization tools. According to
the study of the comparison results, the classification algorithms
with the best accuracy is the Random Forest, with 97.36 %
correctly predicted instances. In a Moodle environment, the
classification techniques might be used to predict students’
performance.