Story generation is the problem of automatically selecting a sequence of events that meet a set of criteria and can be told as a story. Story generation is knowledge-intensive; traditional story generators rely on a priori defined domain models about fictional worlds, including characters, places, and actions that can be performed. Manually authoring the domain models is costly and thus not scalable.
We present a novel class of story generation system--called an Open Story Generator--that can generate stories in an unknown domain. Our system, Scheherazade, (a) automatically learns a domain model by crowdsourcing a corpus of narrative examples and (b) generates stories by sampling from the space defined by the domain model.
Scheherazade can also be used to create interactive narratives in which a player gets to choose the actions for a particular character in the crowdsourced story world. For more on how automatically generated interactive narratives can be used for training and education see the DARPA Press Release.