Cooperative interactions: An exchange values model

Abstract

In non-economic cooperative applications with resource constraints, explicitly motivating cooperation is important so that autonomous service providers have incentives to cooperate. When participants of such applications have different skills and expectations over services, it may be that an agent receives less than expected from a cooperation. A decision-making strategy over interactions in this context must consider not only the motivation to cooperate, but also which interactions to perform to cope with resource limitations. In this paper, we present a computational approach for modelling non-economic cooperative interactions based on the theory of exchange values. Here, exchange values are used to motivate cooperative interactions, and to allow agents to identify successful and unsuccessful cooperations with others, in order to limit service provision and to improve the number of successful interactions. We also present a scenario in which agents participate in a cooperative application in the bioinformatics domain, and show how agents can improve their interactions using the proposed approach. © 2007 Springer-Verlag Berlin Heidelberg.