15th European Conference on Artificial Intelligence
|July 21-26 2002 Lyon France|
The importance of saliency and attention in natural language generation is often underestimated. This paper aims to demonstrate why it is necessary to view and operationalise saliency in a similar way as has happened with intention. The discussion draws on both theoretical and practical considerations, and describes a generation system which implements the model.
Keywords: Discourse Modelling, Natural Language Processing
Citation: Chris Reed: Saliency and the Attentional State in Natural Language Generation. In F. van Harmelen (ed.): ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2002, pp.440-444.