Exchanging Reputation Values among Heterogeneous
Agent Reputation Models: An Experience on ART Testbed Anarosa A. F. Brandão Laboratório de Técnicas Inteligentes — EP/USP, Av. Prof. Luciano Gualberto,
158, trav. 3, 05508-970, São Paulo – Brazil, +55 11 3091 5397 Laurent Vercouter Ecole Nationale Supérieure des Mines de Saint-Etienne, 158, cours Fauriel, 42023,
Saint-Etienne Cedex 2, France Sara Casare Laboratório de Técnicas Inteligentes — EP/USP, Av. Prof. Luciano Gualberto,
158, trav. 3, 05508-970, São Paulo – Brazil, +55 11 3091 5397 Jaime Sichman Laboratório de Técnicas Inteligentes — EP/USP, Av. Prof. Luciano Gualberto,
158, trav. 3, 05508-970, São Paulo – Brazil, +55 11 3091 5397 ABSTRACT

In open MAS it is often a problem to achieve agents' interoperability. The heterogeneity of its components turns the establishment of interaction or cooperation among them into a non trivial task, since agents may use different internal models and the decision about trust other agents is a crucial condition to the formation of agents' cooperation. In this paper we propose the use of an ontology to deal with this issue. We experiment this idea by enhancing the ART reputation model with semantic data obtained from this ontology. This data is used during interaction among heterogeneous agents when exchanging reputation values and may be used for agents that use different reputation models.

 pdflogo.jpg AAMAS07_0187_484617f1fc15aa6b46a67f0ac5e6716b