No abstract

Abstract. Belief-Desire-Intention (BDI) is a well-known cognitive theory, especially in the field of Software Agents. Modelling characters using software agents has been proven to be a suitable approach for obtaining emergent and autonomous behaviours in Interactive Storytelling. In this paper it is claimed that an effective extension of previous models to the BDI framework is useful for designing intelligent characters. An example shows how internal thoughts and motivations of Madame Bovary's main characters can be more naturally formalised as a cognitive side of the story. A narrative reformulation of BDI theory is needed to avoid the implicit complexity of other proposals.

Abstract. The Interactive Dilemma is the inevitable conflict between author's determinism and interactor's freedom. There are some approaches that try to tackle it, using strategies and heuristic rules that can combine on the fly the previously designed author material with the run-time decisions of the interactor. Interactive Narrative is a relatively new field and it is difficult to find formal studies that shows how to create this kind of art. Our proposal is based on the theoretical study of tabletop Role-Playing Games and it involves the practical implementation of those ideas for managing the interaction in a simple text adventure game. Game Masters are the best models we have found in real life for designing and directing interactive stories. In this paper we transfer their player modeling, their rules for interaction management and their improvising algorithms from the real world to a new Interactive Storytelling system.

This paper presents an application that automatically generates basic stories: short texts that only narrate the main events of the plot. The system operates with a representation in Description Logics, combining stored fabulas with the narrative knowledge implemented in a domain-specific ontology. The domain of application is the traditional folk tale, using the wellknown morphology of Vladimir Propp as narratological background. In order to evaluate the results, human judges blindly compared one of the generated basic stories to two alternatives: one rendered directly from a stored fabula of the knowledge base and another randomly generated. As a conclusion, possibilities of measuring the utility of the system in terms of quality and originality of the generated artifact are discussed.

No abstract