Towards data-driven drama management : issues in data collection and annotation

Abstract

One of the key questions in the design and development of interactive drama is structuring an experience for participants such that an engaging, coherent narrative is presented while enabling a high degree of perceived meaningful interactivity. This paper proposes a new approach to the design of intelligent drama managers (DMs) where DM strategies are learned from a corpus of data collected from pen-and-paper RPG game sessions with expert human game masters. In particular, this paper focuses on the issues relating to the collection and annotation of relevant data from recorded gameplay sessions.