Threshold Value in Automatic Learning Style Detection

Abstract

Previous studies demonstrate that automatic approach is a better approach to identify learning styles in online learning because it is based on the actual students’ behavior pattern while learning. However, the threshold value for every behavior pattern has to be obtained to identify students’ learning style. The purpose of this study is to determine the threshold value based on patterns of behavior for 20 students who studied the Interactive Multimedia course. Patterns of behaviour were analyzed based on students’ online learning activities log file. The findings indicate that the threshold values are different depending on for the course structure, subject and experience of different students. Therefore, the determination of the threshold needed to ensure the measurement dimensions of online learning is based on actual usage of the investigated features of the course.