Evolving stories: Grammar evolution for automatic plot generation

摘要: In this paper, we propose a computational framework for automated story-based scenario generation. Under framework, regular grammar is developed to model various causal relationships inside given story world. The then evolved using evolutionary computation techniques generate novel plots, i.e. scenarios. To evaluate these newly generated scenarios, human-in-the-loop used. An experimental study was carried out, in which the proposed used create plots based on famous Little Red Riding Hood fairy tale. demonstrated that can potentially contribute significantly generations. Some challenges were identified including difficulty quantify such subjective measures as plot interestingness and creativity.