Generation of Adaptive Dilemma-Based Interactive Narratives

Abstract

The generator of adaptive dilemma-based interactive narratives (GADIN) presented in this paper dynamically generates interactive narratives which are focused on dilemmas to create dramatic tension. The system is provided with knowledge of generic story actions and dilemmas based on those clichE¿s encountered in many storytelling domains. The domain designer is only required to provide domain-specific information, for example, regarding characters and their relationships, locations, and actions. A planner creates sequences of actions that all lead to a dilemma for a character (who can be the user). The user interacts with the storyworld by making decisions on relevant dilemmas and by freely choosing their own actions. Using this input, the system chooses and adapts future storylines according to the user's past behavior. Previous interactive narrative systems often have content creation and ordering requirements that restrict the possibility for sustaining the dramatic interest of the narrative over a long time period. In addition, many of these systems are not easily transferable between domains. In this paper, the GADIN system is demonstrated to both be able to maintain the dramatic interest of generated narratives over a long time period and to have a core architecture that is applicable to any domain.