Using educational specifications and standards for hypermedia system adaptation according to bcm-inferred student’s learning style

Abstract

A lot of approaches have been developed for adaptation of learning objects of various types and formats to support student-centric learning. Adding adaptivity to personalize learning according to how student perceives, processes, stores, recalls and expresses learning material enables learner to master learning content more effectively and to reach learning goals using attractive ways for a student. Psychologists have come up with many classifications of learning styles. Well known learning style models are already being used in adaptive hypermedia and tutoring systems. Use of data mining, machine learning, case based reasoning and neural networks for automatic learning style modelling continuously improves learning style models, making them more accurate and useful. As learning style models evolve, new approaches for integration of these models with virtual learning environment emerge. Thus, the paper presents an approach for learning hypermedia system adaptation according to students’ learning style inferred by Bayesian case model. Use of Bayesian Case model results in quantitative advantages in learning style prediction quality and interpretability. An approach uses the work of the past to ensure automatic adaptation using educational specifications and standards and tries to apply an enhanced version of it suitable for adaptation to learning styles identified using exemplar-based model.