A Multi-agent Based Adaptive E-Learning System

Abstract

In this paper, a multi-agent based adaptive e-learning system that supports personalization based on learning styles is proposed. Considering that the importance of distance education has increased with the effect of the Covid-19 pandemic, it is aimed to propose an adaptive e-learning system solution that offers more effective learning experiences by taking into account the individual differences in the learning processes of the students. The Felder and Silverman learning style model was used to represent individual differences in students’ learning processes. In our system, it is aimed to recommend learning materials that are suitable for learning styles and previous knowledge levels of the students. With the multi-agent based structure, an effective control mechanism is devised to monitor the interaction of students with the system and to observe the learning levels of each student. The purpose of this control mechanism is to provide a higher efficiency in the subjects the students study compared to non-personalized e-learning systems. This study focuses on the proposed architecture and the development of the first prototype of it. In order to test the effectiveness of the system, personalized course materials should be prepared according to the learning styles of the students. In this context, it is planned to use the proposed system in future studies within the scope of a course in which the educational content is personalized.