Semantic Knowledge Representations for Soft Data Fusion - Efficient Decision Support Systems: Practice and Challenges from Current to Future - page 27

the same real world situation or not, and the information synthesis phase, where compatible
observations of a single real situation are fused.
Regarding situation modeling, we showed that the conceptual graphs formalism could be
used in order to represent situations of interest that have to be monitored.
The association phase relies on the use of similarity measures between graphs structures.
Some parts of the measures are generic whatever the application domain is. Other
components must be customized either by using specific similarity measures, or thanks to
thresholds and weights. The measures we propose take into account the similarity of the
values or referents of the concepts.
The information synthesis phase relies on the use of the maximal join operation defined on
the conceptual graphs structures. We adapted this operation, that was initially proposed by
John Sowa in Sowa (1984), by relaxing the constraints during the similarity testing of two
concept nodes in the fusion process. Through this information synthesis step, we tackle the
problem of soft data fusion and take into account the issue of discrepancies between the
different information items. We use both a compatibility testing and a fusion functions inside
the maximal join operation.
Finally, we show the usefulness of our proposition within a real application. We described
how we propose to take the advantage of information fusion within a TV program
recommendation system.

8. References
Bosse, E., Roy, J. and Wark, S. (2007). Concepts, Models, and Tools for Information Fusion , Artech
House Publishers.
Buford, J., Lewis, L. and Jakobson, G. (2008). Insider threat detection using situation-aware
mas, In Proceedings of the 11th International Conference on Information Fusion , Cologne,
Germany, pp. 212-219.
Chein, M. and Mugnier, M.-L. (1992). Conceptual graphs:

fundamental notions, Revue

d'Intelligence Artificielle 6(4): 365-406.
Chein, M. and Mugnier, M.-L. (2008). Graph-based Knowledge Representation: Computational
Foundations of Conceptual Graphs , Advanced Information and Knowledge Processing,
Springer.
de Chalendar, G., Grau, B. and Ferret, O. (2000). Conceptual graphs generalization, RFIA 2000.
12ème Congrès Francophone AFRIT-AFIA , Paris.
Endsley, M. R. (1995). Toward a theory of situation awareness in dynamic systems, Human
Factors: The Journal of the Human Factors and Ergonomics Society 37: 32-64(33).
URL: http://www.ingentaconnect.com/content/hfes/hf/1995/00000037/00000001/art00004
Gandon, F., Corby, O., Diop, I. and Lo, M. (2008). Distances semantiques dans des applications
de gestion d'information utilisant le web semantique, Proc. Workshop Mesures de
similarites semantique, EGC , INRIA Sophia Antipolis - Mediterranee.
Godbillon-Camus, B. and Godlewski, C. J. (2007). Credit risk management in banks: Hard
information, soft information and manipulati, Munich Personal RePEc Archive.
URL: http://mpra.ub.uni-muenchen.de/
Hall, D. and Llinas, J. (2001). Handbook of Multisensor Data Fusion , CRC Press.
Laskey, K. (2008). Probabilistic ontologies for knowledge fusion, In Proceedings of the 11th
International Conference on Information Fusion , Cologne, Germany, pp. 1402-1409.
Laudy, C. (2010). Introducing semantic knowledge in high level information fusion , PhD thesis,
Université Pierre et Marie Curie.