Semantic Web Technologies Applied to Interoperability on an Educational Portal

This paper describes an approach to promote interoperability among heterogeneous agents that are part of an Educational Portal (PortEdu). We focus on a specific agent, the social agent, adding all the necessary functionality for him to interact with agents that aren’t fully aware of its context. The social agent belongs to a Multi-agent Learning Environment designed to support training of diagnostic reasoning and modeling of domains with complex and uncertain knowledge, AMPLIA. The knowledge of the social agent is implemented with Bayesian networks, which allows the agent to represent its probabilistic knowledge and make its decisions. However, to communicate with agents outside AMPLIA, it is necessary to express such probabilistic knowledge in a way that all agents may process. Such requirement is addressed using OWL, an ontology language developed by W3C to be used on the Semantic Web.