R. Lamarche-Perrin, Y. Demazeau and J.-M.Vincent. « Multi-resolution Representations of Media Information. » In Video Competition of the 23rd International Joint Conference on Artificial Intelligence (IJCAI’13), Beijing, China, August 2013.
ABSTRACT. The analysis of large-scale systems require abstraction tools in order to work at a macroscopic level of description. Data aggregation provides such abstractions by reducing the complexity of the microscopic description. Since it leads to an information loss, such a key process may be extremely harmful for the analysis if poorly executed. This video presents measures inherited from information theory to evaluate abstractions and to provide the experts with feedback regarding the quality of generated descriptions. The aggregation technique is applied to the spatial and temporal aggregation of media information for the analysis of international relations. The approach is able to build multi-resolution descriptions of the system in order to provide efficient and meaningful abstractions to the domain expert.
Voir la vidéo sur le site d’IJCAI 2013.