David Thue and Vadim Bulitko
University of Alberta
While traditional applications of AI in interactive entertainment focus primarily on ensuring that agent behaviours are "good enough", AI techniques can additionally be used to improve the gameplay experiences of players on an individual basis. In this talk, I will present PaSSAGE (Player-Specific Stories via Automatically Generated Events), an interactive storytelling system that automatically learns the preferences of its players to dynamically select the content of an interactive story, toward maximizing their feelings of fun while playing. Following a description of the system's design, I will summarize the results of our empirical evaluations to-date and suggest potential directions for future work.