An automated system based on incremental learning with applicability toward multilateral negotiations

In this paper we propose a negotiation agent system based on the incremental learning in order to increase the efficiency of bilateral negotiations and to improve the applicability toward multilateral negotiations. For the proposed system, we also introduce a framework for multilateral negotiations in an e-marketplace in which the components can dynamically join and disjoin. In order to evaluate the performance of the proposed system, the bilateral negotiation systems based on the trade-off mechanisms have been implemented, and we have extended the systems so that they can perform multilateral negotiations. The experimental results show that the proposed system achieves better agreements than others except for the system under the ideal assumptions that one party knows the personal negotiation information of the other party. Furthermore, the system proposed in our paper carries out negotiations at least twice faster than other negotiation systems implemented in this paper.