Capturing Students’ Dynamic Learning Pattern Based on Activity Logs Using Hierarchical Clustering

Students can have various characteristics and learning patterns. By understanding the pattern of individual students, teachers provide individualized strategies based on students' needs. Students' patterns may experience changes depending their conditions during process. If analysis is only run once, then progress in student throughout process cannot be recognized. On other hand, periodical expected to describe dynamics from time time. This research intended for capturing dynamic using Hierarchical Clustering. We clustered Learning Management Systems (LMS) activity logs. The log data were partitioned into several datasets. results periodic clustering indicated that students’ varied one another changed Most students experienced change semester. also has potential improved maintained.