In this thesis we propose a framework for a system that dynamically selects and plays multimedia files from a large data repository in order to produce a presentation. The presentation is generated based on the technical, semantic and relational textual annotation of the data as well as context-sensitive rules and patterns of selection discovered with the aid of the system during the preparation phase. We borrow concepts from the fields of discourse analysis and rhetorical structure as the theoretical basis of our work. To validate the framework, a prototype was developed using Java, Flash-MX and XML with data created and annotated by a research group from the Department of Design Art.