Context-aware mobile learning on the semantic web / by Xiaoyun Zhang.

Abstract

Progress made in Semantic Web technologies and Ubiquitous Computing has lead to the development of mobile learning services that can adapt to the learner's background, learner's needs, and surrounding environment. In particular, the emerging techniques from these two technologies have the potential to revolutionize the way mobile learning services available on the web are discovered, adapted, and delivered according to context. Context acquisition and management, conceptual knowledge modeling and reasoning, and adaptive services discovery are the main ingredients for designing such context-aware mobile learning systems. However, a number of challenges are still facing the research community in this field. These can be summarized in the following: (i) current mobile learning services act as passive components rather than active components that can be embedded with context awareness mechanisms, (ii) existing approaches for service composition neglect contextual information on surrounding environment, and (iii) lack of context modeling and reasoning techniques for integrating the various contextual features for better personalization. In this thesis an attempt is made to solve the above-mentioned problems. These challenges are addressed by proposing a personalized mobile learning system based on a global ontology space to aggregate and manage context information related to the learner, the used device, the surrounding environment, and the task at hand. The system adopts a unified reasoning mechanism, around the global ontology space, in order to adapt the learning sequence and the learning content based on the learner profile and the perceived contextual information. The adopted approach for ontology reasoning aims at achieving two types of adaptations--system-centric adaptation and--learner-centric adaptation. These are implemented on a Run-Time Environment that identifies new contextual changes and translates them into new adaptation constraints. We developed and tested our system on a number of subject-domain ontologies using various learning scenarios, and the obtained experimental results are very promising.