论文部分内容阅读
为提高基于智能体(agent)的电子商务多边多议题协商的效率及稳定性,提出改进的自适应差分进化算法(ADE)并将其引入到合作环境下的多边多议题协商问题中.差分进化(DE)算法是目前求解连续空间内全局优化问题性能最优的进化优化算法之一.利用该算法收敛速度快、收敛精度高、全局寻优能力强等特点加快多边多议题协商的速度,使协商效率更高、稳定性更强.通过与目前解决多边多议题协商问题效果最好的混合遗传算法(HGA)对比,实验结果表明,自适应差分进化算法具有更快的收敛速度和更好的稳定性,可以使多边多议题协商中的各智能体达到协商最优解,并有效地减少协商次数,提高协商的效率和稳定性.
In order to improve the efficiency and stability of agent-based multilateral multi-issue e-commerce negotiation, an improved adaptive differential evolution algorithm (ADE) is proposed and introduced into the multi-agent multi-issue negotiation problem under cooperative environment. (DE) algorithm is one of the evolutionary optimization algorithms to solve the optimal global optimization problems in continuous space at present. This algorithm accelerates the speed of multilateral multi-topic negotiation based on its fast convergence speed, high convergence precision and global optimization ability The negotiation is more efficient and more stable.Compared with HGA, which is the most effective way to solve the multi-lateral multi-issue negotiation problem, the experimental results show that the adaptive differential evolution algorithm has faster convergence speed and better Stability can make all agents in the multilateral multi-topic negotiation achieve the optimal solution and reduce the number of consultations effectively, so as to improve the efficiency and stability of the negotiation.