网络结构和有界信任对群体观点演化过程的交互影响  

The Reciprocal Effect of Network Structures and Bounded Confidence on Group Opinion Evolution

作　　者：宋艳双[1] 刘人境[1] 

机构地区：[1]西安交通大学管理学院,西安710049

出　　处：《软科学》2016年第1期120-123,144,共5页Soft Science

基　　金：国家自然科学基金项目(71271166)

摘　　要：对HK模型进行扩展,综合考虑不同观点之间的影响权重以及交互网络的结构特征,探讨了在有界信任和影响异质性的前提下,网络结构和信任阈值对群体观点演化结果的交互影响。仿真结果表明,随着有界信任参数的增加,不同网络结构类型下的平均观点簇数量均呈现出递减的趋势,且递减速度受网络结构的影响。小世界网络随着重连概率的增加,平均观点簇数量的下降趋势更明显,并且群体观点演化速度不断加快,使得网络能够在较短的时间内形成更大的观点簇,从而更有利于群体共识的达成。规则网络中观点的收敛速度最慢,最终网络中保持着较高的观点簇数量,使得群体观点比较分化,一致性构建的难度加大。Considering the influence weight of different opinions and the structures of interacting networks,this paper extends the HK model and discusses the reciprocal effect of network structures and bounded confidence parameter on group opinion evolution under the premise of bounded confidence and influence heterogeneity. The simulation results show that,as the bounded confidence parameter increases,the average opinion clusters under different network structures will decrease,but the speed is different in different network structures. As the rewiring probability increases,the downward trend of average opinion clusters in small world network is more obvious than other networks,and also the rounds required to reach a convergent state is fewer,making the network can form a bigger opinion cluster in a relatively shorter period of time. Regular network requires longest time to reach a convergent state,and final network maintains a higher number of opinion clusters,making group opinion more differentiation and harder to build global consensus.

关 键 词：HK模型 有界信任 网络结构 群体观点演化 

分 类 号：F713[经济管理—产业经济] C934[经济管理—管理学]