Applied clustering analysis for grouping behaviour of e-learning usage based on meaningful learning characteristics

Abstract

One of the critical success factors of e-learning is positive interest of students towards e-learning. The majority of activities of current e-learning usage are viewing and downloading. These activities are not meaningful with regard to enhancing learning quality. Due to that, the aim of this paper is to analyze students’ usage based on meaningful learning characteristics by clustering students’ activities and actions during online learning. We first define meaningful learning characteristics (as those which are active, authentic, cooperative, collaborative, and intentional) and associate these with e-learning activities and actions. Then, we analyze the students’ e-learning usage and define the cluster of student’s meaningful characteristics by using the K-Means cluster method. A case study has been conducted based on the e-learning log files of 37 students on Computational Intelligence Course at the Software Engineering Department, Universiti Teknologi Malaysia. The result of this clustering enables us to determine the students with high ratings on these meaningful activities and actions during online learning. We found out that students with high hits on add, update, and edit are included in the high cluster group. On the contrary, students with high hits on the view actions for all e-learning activities are included in the low cluster group. This result may assist instructors while preparing the strategy of computer usage for education, in terms of providing a greater variety of learning activities, which is applicable for any courses.