The emergence of Web of Data enables new opportunities for relating resources identified by URIs combined with the usage of RDF as a lingua franca for describing them. There have been to date some efforts in the direction of exposing learning object metadata following the conventions of Linked Data. However, they have not addressed an analysis on the different strategies to expose Linked Data that could be used as a basis for leveraging the metadata currently curated in repositories following common conventions and established standards. This paper describes an approach for exposing IEEE LOM metadata as Linked Data and discusses alternative strategies and their tradeoffs. The recommended approach applies common principles for Linked Data to the specificities of LOM data types and elements, identifying opportunities for interlinking exhaustively. A case study and a reference implementation along with an evaluation are also presented as a proof of concept of this mapping.