Semantic Web Approach to Content Personalization

Our environment is complex and the amount of information is exceeding any volume being manually manageable by humans. Information needs to be adjusted and personalized to be of avail to each individual user. In this paper, a novel software system is described achieving content personalization with high quality and in an automatic manner. The approach taken for creating new personalized services is based on the synergetic effects through the combination of semantic Web, AI and P2P technologies. Although various research projects examine these technologies, few currently deal with the issues to unify the advantages. Key features are the semantics of RDF metadata and DAML+OIL ontology, applied for enriched description, querying and inference. The learning algorithm ID3, used for revealing user preferences, and Jupiter, a peer-to-peer platform leveraged for mobile semantic applications in a ubiquitous communication environment, were applied. The concept and underlying technical challenges are described and illustrated by a use case, and the details of a prototype implementation are explained.