Automatic Generation of Semantic Metadata as Basis for User Modeling and Adaptation

With the help of the simple and world-wide accepted technique of tagging, users can help to collaboratively provide metadata over previously uncharted collections of multimedia documents. However, the semantics of tags are rather limited and not always as helpful in disclosing a dataset as a proper ontology can be. In this paper we introduce the Relco framework that applies syntactic, semantic and collaborative techniques to connect tags to ontological concepts, which helps to quickly get more semantics about a tag. We demonstrate the applicability of our techniques in two concrete Web applications: one in the educational domain and one in the cultural heritage domain. For the former we describe how students are better able to find the information in socially tagged videos and in the latter we also show how the used techniques allow building a faceted browser over the previously uncharted multimedia objects and we show which techniques could be applied to control the quality of user driven annotation.