Architecture d'agent sensible au contexte pour la prise de décisions

ABSTRACT

Sensing, managing and representing context are necessary capacities for ubiquitous computing applications, moreover, their components must also be able to reason and adapt their behavior to the evolution of their physical and social environment. This paper proposes a multi-agent architecture consisting of context aware agents able to learn how to discern relevant from non relevant context on one hand, and to make appropriate decisions based on it on the other hand. This multi-agent system interacts with a context manager layer, which is able to answer context-related queries by using an ontological representation of context. The use of this architecture is illustrated on a test application for agenda management, using the JADE-LEAP platform on PCs and PocketPCs.

REFERENCES

• 1: Bucur O, Boissier O, Beaune P - "A Context-Based Agent Architecture for Learning How to Make Contextualized Decisions" (to appear in Proc. of MCMP 05), may 2005.
• 2: Harry Chen , Tim Finin , Anupam Joshi, An ontology for context-aware pervasive computing environments, The Knowledge Engineering Review, v.18 n.3, p.197-207, September 2003 [doi>DOI:10.1017/S0269888904000025]
• 3: Data Mining II: http://www.comp.nus.edu.sg/~dm2/
• 4: Dey, A., Abowd, G.- "Towards a better understanding of Context and Context-Awareness", GVU Technical Report GIT-GVU-00-18, 1999.
• 5: JADE: http://jade.cselt.it/
• 6: Jena: http://jena.sourceforge.net/
• 7: Lashkari Y. et al - "Collaborative Interface Agents", Proc. of the Third International Conference on Information and Knowledge Management CIKM'94, ACM Press, 1994.
• 8: Lin S., J. Y. Hsu - "Learning User's Scheduling Criteria in a Personal Calendar Agent", Proc. of TAAI2000, Taipei.
• 9: Tom M. Mitchell , Rich Caruana , Dayne Freitag , John McDermott , David Zabowski, Experience with a learning personal assistant, Communications of the ACM, v.37 n.7, p.80-91, July 1994 [doi>10.1145/176789.176798]
• 10: OWL: http://www.w3.org/2004/OWL/
• 11: Protégé 2000 - http://protege.stanford.edu/
• 12: Sen S., E. H. Durfee - "On the design of an adaptive meeting scheduler", in Proc. of the Tenth IEEE Conference on AI Applications, p. 40--46, 1994.
• 13: Paul Scerri , David V. Pynadath , Milind Tambe, Why the elf acted autonomously: towards a theory of adjustable autonomy, Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 2, July 15-19, 2002, Bologna, Italy [doi>10.1145/544862.544944]
• 14: Tao Gu et al - "An Ontology-based Context Model in Intelligent Environments", Proc. of Communication Networks and Distributed Systems Modeling and Simulation Conference, 2004.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
        I.2.11 Distributed Artificial Intelligence

Keywords:
context, learning, multi-agent systems