基于对手案例学习的多议题协商决策模型  

Based on the study of opponent case mutil-issues negotiate decision model

作　　者：范京津[1] 傅明[1] 

机构地区：[1]长沙理工大学

出　　处：《微计算机信息》2011年第9期215-216,233,共3页Control & Automation

摘　　要：为了解决多Agent系统(MAS)协商双方在信息对称情况下的自动协商问题,提出了一种用基于支持向量机算法的间接学习对手协商态度的协商方法,提出了不完全信息条件下基于案例和对策论的Agent多议题Pareto最优协商模型,通过支持向量机的方法来学习协商轨迹,得到协商对手在每个协商项的态度,然后利用学习得到的对手协商态度,构造了一个协商的决策模型,此模型能同时基于对手的态度和自身的偏好来做出协商决策。最后通过实验验证了该方法的先进性。In order to solve the mutil-agent system（MAS） negotiations both sides in the information symmetry under the circumstance of auto-negotiation problem,this paper proposes an algorithm based on support vector machine with the indirect learning attitude negotiation opponent consultation method,puts forward under the condition of incomplete information based on case and countermeasures concerning the Agent Pareto optimal issues more negotiation model,through the support vector machine approach to learning consultation trajectory,get consultation in each negotiation opponent of attitude,and by using the study to get opponent consultation attitude,constructs a negotiated decision-making model,this model could also based on the opponent＇s attitude and their preference to make decision consultation.Finally,the experiments show that the presented method is advanced.

关 键 词：协商态度 支持向量机 多议题协商 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]