ELDO: An Ontology for the Cataloguing of e-Learning Design

Recently there have been many both open and private courseware repositories appearing on the web. In general however, there has been great commitment to building communities that facilitate the liaison between users/suppliers, but quite a modest employment of sometimes very elementary cataloguing systems. When the ILOR project (Italian Learning Object Repository) was launched, which created an Italian LOR experience, CERTE (Research and Development Centre for Elearning Technologies Centre - Omniacom.org) wanted to undertake the rigorous construction of an eLearning Design Ontology (ELDO), starting with university didactic requirements and with the Dublin Core metadata (DCMI) model, to then compare them with the SCORM LOM model and with a careful analysis of the application area. The research has already produced a stable cataloguing model which integrates DCMI, LOM SCORM in a metadata structure called PLOM2 (Pegasus Learning Object Metadata) by Pegasus LMS (R) technology used for implementation. PLOM2 is already used for cataloguing the current repository while awaiting the final ontological model (PLOM3). The research-development cycle described here corresponds with a bottom-up process that was conducted starting from three repertories: the cataloguing of courseware produced for University of Ferrara's distance-learning degrees and Masters, the DCMI base set and the LOM originating from the IMS model, adopted by SCORM. However, the project had already detected potential classification difficulties operating with the bottom-up process only. Many categories in fact (especially associated with users, their work and training organization, and course planning etc.) could not find an adequate placement within the model. An R&D process was therefore envisaged parallel to the top-down modality dedicated to ontology construction. The starting point was once again the university courseware cataloguing requirements. Beginning with these needs, ELDUM (eLearning Design unified model) was designed. This is a model of thematic depiction built from the correlation between the mapping class-concepts and their correlation with specific relation-properties. A database was placed alongside ELDUM's dynamic conceptual map that allowed the gradual archiving of both classes and properties. OWL is used as a reference model due to the frequent randomness of definitions, while Protege is used for building the ontology. The objective is to obtain a stable version of ELDO by the end of 2008, in order to provide first PLOR and then ILOR with a stable ontological instrument. Finally it is important to note on the semantic analysis of content side that it is operating with a Pegasus LMS (R) OKM Open Knowledge Management module, for the semantic analysis of texts, the automatic construction of glossary and thesaurus, as well as the automatic marking of the same texts going on to be catalogued.