共演博弈下的网络合作  

Network cooperation of co-evolutionary games

作　　者：仇建平[1] 

机构地区：[1]太原科技大学计算机科学与技术学院,山西太原030024

出　　处：《智能系统学报》2014年第1期88-91,共4页CAAI Transactions on Intelligent Systems

基　　金：山西省回国留学人员科研基金资助项目(2013-097)

摘　　要：针对社会网络上个体之间的竞争与合作,设计了社会网络模型。通过引入博弈理论,仿真社会网络个体间复杂的交互关系,即合作者或背叛者依据收益矩阵和邻居节点进行博弈。仿真结果显示当影响度过低或过高时,合作不易形成,影响度适中时合作水平最高;当收益固定时,合作变为扩散,形成更高的威望。最后,通过MovieLens数据集验证了群聚效应的存在,分析了舆论领袖对于网络群体的巨大影响力。In accordance with competition and cooperation among individuals in social networks , a social network model has been designed .Game theory is introduced to simulate the complicated interaction relations among individ-uals in social networks .Cooperators or defectors interact with their neighbors following the payoff matrix .It can be seen that the cooperative behavior is easy to extinct for too low or too high values of degrees of influence , and the maximum cooperation level is achieved for medium values of degrees of influence .When the reward is fixed , the co-operative behavior can gradually diffuse and build higher reputations .Finally, we verified the existence of the clus-ter effect based on the MovieLens dataset and analyzed important influences of opinion leaders in netizen groups .

关 键 词：社会网络 共演博弈 合作 群聚效应 舆论领袖 

分 类 号：O157.5[理学—数学]