•  
  •  
 

Abstract

Learning Management Systems (LMSs), such as Blackboard, have been shown to be valuable for online teaching and learning activities. They offer many tools for students and instructors. Moreover, they can track and record users’ activities, which can be analyzed using data mining techniques. The success of LMSs depends heavily on instructors because they typically initiate the activities and encourage the students to use LMSs. However, several studies indicate that instructors’ use of LMS features is often limited. This study aimed to identify the patterns of instructors’ usage of Blackboard features using a clustering technique. The experiment was conducted at King Abdulaziz University, using data obtained from the Blackboard database from the first semester of the 2020 academic year. We analyzed the data from 11,667 course sections and 20 colleges, comparing three clustering techniques: self-organizing map, k-means, and hierarchical. The results of the comparison revealed that the self-organizing map outperforms the other algorithms. Subsequently, we analyzed the resulting clusters to identify usage patterns. We identified five distinct usage patterns: inactive, adding contents and testing, evaluations, communicative, and balanced. Additionally, we identified the relations between the usage patterns and colleges.

Keywords

Blackboard, data mining, hierarchical clustering, k-means, learning management system, SOM clustering

Article Type

Article

First Page

63

Last Page

73

Publication Date

4-30-2025

Download

Share

COinS