Automatic Detection of Learning Styles on OpenEdx Using Trace Analysis

Abstract

In a learning management system (LMS), the tutor does not interact with the students in person to interpret their reactions and grimaces or to determine the level of course assimilation. As a result, analyzing learners' behavior and traces in an e-learning environment has become necessary for tutors or teachers to assist their students. Many researchers argue that learners' learning styles are an important factor to consider in LMS. These learning styles enable students to tailor their learning to their specific environment. Every learner has a unique learning style and way of perceiving, processing, retaining, and comprehending new information. This approach, which is based on the analysis of learners' interaction traces in LMS, gives teachers a sense of their students' behavior and identifies their learning styles. Our method has demonstrated an accuracy rate of 80% in detecting learning styles, significantly enhancing engagement and motivation. By tailoring content, activities, and assessments to match each student's preferences, we can adapt interventions dynamically.