Development of a Multi-issue Negotiation System for E-Commerce

Agent-mediated e-commerce is rapidly emerging as a new paradigm to develop business intelligent systems. Such systems are built upon the foundations of agent technology with a strong emphasis on the automated negotiation capabilities. In this paper, we address negotiation problems where agreements must resolve several different attributes. We propose a one-to-many multi-issue negotiation model based on the Pareto optimal theory. The proposed model is capable of processing agents’ preferences and arriving to an optimal solution from a set of alternatives by ranking them according to the score that they achieved. We present our prototype system architecture, together with a discussion of the underlying negotiation framework. We then illustrate on our implementation efforts using the JADE and Eclipse platform. Our concluding remarks and possible further work are presented.