Understanding what is going on, or how to make sense of situations

Philippe Chassy (1) , Didier Dubois (2) , Henri Prade (2)

1 Liverpool Hope University
2 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage

Understanding what is going on in a sequence of reported events is of interest in a variety of tasks ranging from surveillance problems to the modeling of how people may perceive the information with which they are faced. Generally speaking, it is a complex task that in- volves the identication of relevant patterns involving abnormal events, and requires the use of background knowledge. The notion of causality plays a central role in such problems, when one has to decide which events are bound by a causal relation (while other events are irrelevant with respect to the focus of interest), and maybe to predict what is likely to take place. The paper restates a model of causality ascription that has been recently proposed, and suggests how more generally the consonance or the dissonance between what is reported and knowledge about the normal course of things impact the way one can make sense of a situation. Some other related issues are brie y mentioned in the concluding remarks.

Domaines: Intelligence artificielle [cs.AI]

Philippe Chassy, Didier Dubois, Henri Prade. Understanding what is going on, or how to make sense of situations. 1st International Workshop on Uncertainty Reasoning and Multi-Agent Systems for Sensor Networks (URMASSN 2011), Jun 2011, Belfast, Ireland. ⟨hal-03351548⟩