Concept Maps and Course Ontology: a Multi-level Approach to E-learning

The work is mainly focused on the description of an original scheme for information representation in e-learning applications, where three levels are used to describe course related knowledge: content level, symbolic level, and conceptual level bridging the previous ones. Moreover, an integrated architecture design is proposed to support the presented model. The reference scenario is the development and deployment of lessons for undergraduate students using both local and internet resources. Multimedia information is managed at a content level relying on the SCORM standard. At a symbolic level, an ontology is built as a linguistic knowledge base, where all the course topics are represented together with temporal and user-related constraints. The ontology is used to generate a suitable learning path in response to the student requests, which has the form of a SCORM course description. At the intermediate level, which bridges the content and the symbolic ones, topics are represented by a concept map, implemented as a SOM network. The map is used to cluster the course materials, and to map them onto atomic concepts that can be instantiated at the ontological level. This solution allows the discovering similarity between topics, and creating new relations between them at the symbolic level. A detailed description of the model is presented, along with the outline of the architecture.