Semantic interoperability model for learning object repositories

Abstract

Interoperability among repositories requires syntactic and semantic compatibility, achieved through the adoption of metadata standards. However, different learning object repositories currently use diverse metadata standards to describe their resources, so multiple metadata standards describe the same term, and vice versa. This paper proposes an ontologybased interoperability model, featuring a shared vocabulary (SV) and a set of matching rules. The SV establishes a common terminology for learning objects, while the matching rules enable translation between the SV and any metadata standard. As a result, both deposit and search for learning objects can be conducted using any metadata standard, thanks to the rules that ensure seamless translations where needed. To evaluate the proposed model, a prototype has been developed, which implements the SV and matching rules. This experience has shown that using ontologies and matching rules to provide an interoperability model for learning objects repositories is a valid, flexible and user-friendly solution.