Abstract |
Online courses, web-based education, computer supported training and even virtual university are already wide used terms. All of them represent e-learning which is growing very fast both in educational and corporate environment. In particular, e-learning systems that are offered via the world-wide web can be considered as specific web-based information systems with focus on the provision of knowledge to learners. Several challenges have to be faced in order to facilitate an open dynamic e-learning environment. /Incomplete/ knowledge and /conflicts/ are two main issues emerging in this context. We should also stress that /personalization/ and /recommendation/ tasks further imply the management of incomplete knowledge. Moreover, decisions, concerning personalization/recommendation, made at a time may become invalid later, after the consideration of a new piece of knowledge. In this work we elaborate on the design and implementation of a personalized rule-based e-learning system using Semantic Web technologies. In particular, a potential user of our system can navigate between personalized Web pages as well as to his/her knowledge level. The learner knowledge level for each subject is deduced by the reasoning module. This module uses logic over online RDF descriptions, to conclude or guess the user knowledge level. The reasoning module makes also recommendation to the learner recommending the most appropriate content to focus his attendance. Additionally, for reasoning with inconsistent or incomplete information, which is a common phenomenon, we use defeasible logic. Its nonmonotonic behavior supports easy revision of system hypothesis about user knowledge when data is considered, without having inconsistencies. To the best of our knowledge, our work is the first/ combining the advantages of Semantic Web with Defeasible Logic reasoning in the domain of e-Learning/.
|