作者: Boyang Li , Mark O. Riedl
DOI:
关键词:
摘要: Most computational story generation systems lack the ability to generate new types of imaginary objects that play functional roles in stories, such as lightsabers Star Wars. We present an algorithm generates objects, which we call gadgets, order extend ontological expressivity existing, planning-based systems. The behavior a gadget is represented plan including typical events happen when used. Our creates gadgets by extrapolating and merging one or more commonly known achieve narrative goal provided existing generator. partial-order planning establish open conditions based on analogies between concepts related respectively common gadget. show capable generating created human.