Pablo Gervás Gómes-Navarro: Modelling How People Build Stories from Observed Facts

Notice: This particular AI Agora event will NOT be taking place at Media City Bergen. This time, the venue for the event is the Social Science Building, Fosswinckelsgt. 6. Please see details below.

Abstract

People communicate most often by telling stories based on their experience of reality. In this process, three fundamental ingredients need to be combined: the observed facts as perceived from reality, the knowledge of narrative structure that is shared between the teller and the listener, and the particular purpose that drives the teller to communicate. Depending on the purpose the story that results will be closer to the facts on which it is based (if the main purpose is to report them) or loosely based on them and adorned with invented details or even fictional events (if these additions make it better serve whatever purpose is intended, such as entertaining or exercising our sensibility).

Narrative communication can then be seen to operate between two different modes. In both cases, discourses satisfying the requirements of narrative structure are produced. However, there are significant diffferences in the way the task of deciding what content is to be included in the story (what is known in the field of natural language generation as the content determination strategies employed during composition). Whereas in one case the process is mainly one of selection, in the other there is likely to be a large percentage of invention. Yet both the selection and the invention are likely to be guided or informed by considerations of narrative structure and successful narrative outcomes.

The talk will describe recent work on computational modelling of these two narrative modes and present insights arising from their joint consideration into a single framework capable of switching from one to the other (and considering hybrid modes of operation) much as humans are seen to do.

Biography

Dr. Pablo Gervás holds a PhD in Computing from Imperial College, University of London (1995). He is currently associate professor (Profesor Titular de Universidad) at Universidad Complutense de Madrid. He is the director of the NIL research group and secretary of the Instituto de Tecnología del Conocimiento.

He has been coordinator for two national research projects (GALANTE and MILES) involving several institutions. Dr. Gervás has published research papers in the fields of expert systems, knowledge representation, case-based reasoning, information retrieval, natural language generation, multi-agent applications and Computational Creativity, including papers in journals such as Journal of Knowledge-based Systems, New Generation Computing, Information Processing and Management and Artificial Intelligence Magazine. His main research interest currently lies in the study of the role that computers can play in helping people interested in literary creativity.

Dr. Gervás is one of the world’s leading experts on automatic generation of (fictional) stories and poetry, and has a background in Natural Language Generation, Computational Creativity, and Computational Narratology. He is the author of the PropperWryter software which was used in the process of creating Beyond the Fence — the first computer generated musical, staged at the London West End in 2016.