A learning style-based Ontology Matching to enhance learning resources recommendation

Abstract

Content-based recommender systems (CBRS) are widely used in various domains including the E-leaning. In the E-learning field, a content-based recommender system aims to represent and extract the knowledge about learners and the learning materials to personalize recommendations by determining which learning material fits more the learner's need. The knowledge in a CBRS can be represented using ontologies, since knowledge representation is a vital step in the knowledge-based technique. Knowledge about the learner is represented in the form of a learner's profile that includes a lot of information about the learner such as the learner's level of knowledge, skills, requirements and the learning style which is the learner's characteristic that we have relayed on in our proposed ontology based recommender system for E-learning. In this paper, we have used the learning style, following the model proposed by FelderSilverman (FSLSM) to model the learner profile and the learning content. On the one hand, each learning material is mapped to a specific learning style based on the FSLSM, and on the other hand, the learners' learning styles have been automatically detected using a machine learning-based approach. The learner's profile and the learning content will be modeled using two ontologies, and then the learning style will be used as a common data-type property to match them and generate personalized recommendations.