An Agent Negotiation System Based on Adaptive Genetic Algorithm

In agent bilateral multi-issue bidding negotiation system, how to make the negotiation agents gain satisfying result farthest and negotiate efficiently is a key issue. As for this problem, an agent negotiation system based on adaptive genetic algorithm is present and the algorithm is applied in bilateral multi-issue simultaneous bidding negotiation in E-commerce. In the system, the negotiation agent can send the information of its issue range and the issue weight to the third party agent which it trusts, and then the third party agent use the adaptive genetic algorithm which is present in this paper to give the optimal result. In the system experiments, the two methods are used to compare. The first is simple genetic algorithm (SGA), the second is the adaptive genetic algorithm (AGA). The difference is the later algorithm can change the crossover probability and the mutation probability adaptively. The SGA uses 218 runs to gain the satisfying result, while the AGA only uses 152 runs to gain the satisfying result. The experiments show that the system present in this paper can help agents to negotiation more efficiently.