Semantic Web Model for Personalized E-Learning

Abstract

Personalized e-Learning is a highly interested research area among the e-Learning research community aimed in customizing the learning process of e-Learning based on needs and preferences of the learner. Personalized e-Learning adapt the learning process according to the factor's like learning objects, user prefer- ences, background knowledge, context, user's competency levels, and learning patterns instead of providing a learning model that one-size- t-all like in con- ventional e-Learning. Proper mechanisms to handle content representation and maintaining user pro ling are the main factors to be considered in this context. Exiting solutions for the personalized e-Learning is highly depends on metadata de ned for the content and adapting various standards to model content and user. It is already proven that these kinds of approaches did not su cient enough to provide adequate information for a service like personalization. Proposed solution is focused on a suitable approach to build a semantic web model for personalized e-Learning applying semantic web concepts that can be get rid of mentioned drawbacks. Semantic web concept is a novel concept that allows web content to be expressed not only for humans but also in a format that can be read by the machine. Ontology's under the Semantic Web can be used for knowledge representation and knowledge sharing by modeling concepts and relationships in given domain and Semantic Web have been used for content representation for the past few years. Proposed system suggests a novel mechanism to handle content representation for personalized e-Learning in a more expressive way by emphasizing the impor- tance of maintaining a separate knowledge layer for knowledge representation and content layer to store content according to the knowledge represented and this enables search the knowledge user expects to learn when user has a goal to follow and then present relevant content rather than blindly search the knowledge from the resources using Meta data. Subject domain is modeled using ontology as the knowledge layer for knowledge representation in a way that gives common understanding of the subject domain. Resources would be stored in the content layer according to the knowledge representation. User pro le which is responsible for collecting user information and user feedback and also user assessment that will help to deliver personalized content to the learner is also modeled based on ontology by mainly capturing the features of the user like background level, com- petency and preferences. Personalized needs would be served by the proposed solution in a goal oriented way by mapping domain ontology and user pro le ontology. A Learning path would be generated from the mapping and it would be validated with user's background knowledge and then results in generating the most matching learn- ing path for the learner. Resources would be extracted based on this suggested learning path for the learner.