Semantic Agents for Learning Style Recognition in an e-Learning Environment

This paper aims to provide a survey of the major research works done in the domain of learning style recognition in an e-learning environment and proposes a Semantic Agent Framework for the e-Learning environment to detect individual differences existing among individuals, using their learning styles. Automatic detection of learning styles of an individual in an e-learning environment is an important problem that have been researched upon by many, as it proves beneficial to the learners to be provided with materials based on their individual preferences. To achieve this dynamic adaptability, we propose to use a mix of data-driven approach and literature-based approach. Out of the 71 models of learning styles that are described by different researchers, we consider the Felder-Silverman Learning Style Model (FSLSM) for our analysis.