Laboratoire d'Informatique de Grenoble Équipe Ingénierie de l'Interaction Humain-Machine

Équipe Ingénierie de l'Interaction
Humain-Machine

Towards a computational generation of récit: evaluating the perception of the récit plan

In 1st International Workshop on Data-to-Text Generation. 3 pages. 2015.

Belén A. Baez Miranda, Sybille Caffiau, Catherine Garbay, François Portet

Résumé

The aim for this PhD project is to generate stories about human activities from real ambient data.
Computational Narratology (CN) is the study of narratives from the point of view of computation and information processing (Mani, 2012).
Most of the current researches in CN are related to creativity, where the stories emerge from a set of predefined parameters, trying to imitate literary genres like fairy tales (Riedl and Young, 2010).
In this paper we are interested in stories emerging from real ambient data for which we have no control.
We name these stories ”activity récits”.
One challenge addressed is the coherence between a story plan (i.e. a sequence of events) rendered as text read by humans and real activity performed by actors and gathered through sensors.
This paper aims to present an evaluation of the perception of the story plan from a tool employed to express human activities: a task model sequence.