Adressing context-awareness and standards interoperability in e-learning: a service-oriented framework based on IRS III

Current technologies aimed at supporting learning goals primarily follow a data and metadata-centric paradigm. They provide the learner with appropriate learning content packages containing the learning process description as well as the learning resources. Whereas process metadata is usually based on a certain standard specification – such as ADL SCORM or the IMS Learning Design – the used learning resources – data or services - are specific to pre-defined learning contexts, and they are allocated manually at design-time. Therefore, a content package cannot consider the actual learning context, since this is only known at runtime of a learning process. These facts limit the reusability of a content package across different standards and contexts. To overcome these issues, this paper proposes an innovative Semantic Web Service-based approach that changes this data- and metadata-based paradigm to a context-adaptive service-oriented approach. In this approach, the learning process is semantically described as a standard-independent process model decomposed into several learning goals. These goals are accomplished at runtime, based on the automatic allocation of the most appropriate service. As a result, we address the dynamic adaptation to specific context and - providing the appropriate mappings to established metadata standards - we enable the reuse of the defined semantic learning process model across different standards. To illustrate the application of our approach and to prove its feasibility, a prototypical application based on an initial use case scenario is proposed