A Context-Based Framework And Method For Learning Object Description And Search

Xiaofeng Du,William Song,Ming Zhang
DOI: https://doi.org/10.1007/978-3-540-78139-4_11
2008-01-01
Abstract:For the last decade, E-Learning has become an active research area. Many companies and organisations are now providing large amounts of online learning resources. These learning resources have covered most common education and learning areas and subjects and are always available so that the learners can access them from anywhere which has an Internet connection. Learners can flexibly choose the subjects they want and build up their own curriculum and study schedule. However, most of the online learning resources are poorly described and structured so causing huge problems in their use, search, organization, and management. To overcome the problems, we propose a novel and practical context-based semantic description framework which aims to describe information and knowledge about learning resources and their structures. Context-based semantic description is an effective way to extract knowledge from various aspects to depicting learning resources which are abstracted and termed as "Learning Objects". This framework consists of four parts: the definition of Learning Objects, a Context-based Semantic Description Model, an ontology, and learning concept dependency graphs. By using the Learning Object's attributes and their various semantic relationships addressed in the proposed framework, we attempt to search and match a learner's requirements against the description of Learning Objects provided by the framework with the help of knowledge from learning environments. A key step here is to compute semantic similarity using the modelled knowledge. The proposed work aims to support learners in using massive learning resources from the web to perform self-learning with or without the help of educators' advice and instruction.
What problem does this paper attempt to address?