Script knowledge plays a central role in text understanding and is relevantfor a variety of downstream tasks. In this paper, we consider two recentdatasets which provide a rich and general representation of script events interms of paraphrase sets. We introduce the task of mapping event mentions innarrative texts to such script event types, and present a model for this taskthat exploits rich linguistic representations as well as information ontemporal ordering. The results of our experiments demonstrate that this complextask is indeed feasible.
展开▼
