<![CDATA[The rapid growth of e-learning has increased the need to analyze student engagement levels in online learning environments. This study aims to analyze student activity levels using a clustering method based on activity log data from a Learning Management System (LMS). The approach applied is K-Means Clustering with an exploratory method on student activity data, including frequency of access to learning materials, discussion forum participation, assignment submission, and duration of learning activities. The sample consisted of 200 Informatics Engineering students selected through purposive sampling, with data collected over one academic semester. The optimal number of clusters was determined using the elbow method, while cluster quality was evaluated using the Silhouette Score. The results show that students can be grouped into three main clusters: high, medium, and low engagement levels. The high-engagement cluster showed consistent participation and potential correlation with better academic performance, while the low-engagement cluster indicated a higher risk of academic difficulty. These findings are consistent with previous studies showing that interaction levels are strongly correlated with academic performance. Clustering methods proved effective in automatically identifying student behavior patterns, thereby supporting the development of data-driven adaptive learning strategies.]]>
Copyrights © 2026
