Evaluation of e-learning as a learning agent

Abstract

The aim of this study was to evaluate the e-learning as a learning agent. The main feature of this system is to identify a learning style of the student through a log system that relates the students' achievement in mastering the learning content. The students' learning style was determined by Bayesian network algorithm that enabled the system to classify the students' learning style into textual, audio, or video. With this identification, the system would be able to recommend suitable learning contents for the students. The study revealed that the e-learning was capable of identifying the students' learning style. From the total 34 research subjects involved in the study, the system successfully identified 16 students' learning styles, where 14 students were revealed to have textual learning style, one student was revealed to prefer audio content, and the other student was revealed to prefer video content. Thus, the study concluded that the adaptive e-learning system evaluated in the study was capable of identifying the students' leaning styles, and thus capable of recommending suitable materials for the students, hence rendering the e-learning system as a mere content management system, but also as an agent of learning.