摘要
传统网络舆情演化研究中,网络结构平均聚类系数较小,平均路径长度较大,且个体之间的相对权威性相等。针对这种情况,提出一种基于个体相对权威性的改进HK(Hegsekmann-Krause)模型来对个体间的权威性进行量化,同时构建更符合实际社交网络性质的网络拓扑结构。实验结果表明:该模型随着有限信任阈值增大,演化后的最终观点数量减少;随着平均节点度增大,观点以更快的速度趋于一致;该模型能够有效提高网络平均聚类系数、降低平均路径长度,同时舆情演化效果与传统HK模型一致。
In the research of traditional network public opinion evolution,the average clustering coefficient of the network structure is small,the average path length is large,and the relative authority between individuals is equal.Therefore,this paper proposes an improved HK model based on individual relative authority to quantify the authority between individuals and builds a network topology that is more in line with the nature of the actual social network.The experimental results show that as the bounded confidence threshold increases,the number of final opinions after evolution will be less;as the average node degree of the network increases,the views will tend to be consistent at a faster speed.This model can effectively improve the network average clustering coefficient and reduce the average path length.At the same time,the evolution effect of public opinion of this model is consistent with the traditional HK model.
作者
马永军
柴梦瑶
Ma Yongjun;Chai Mengyao(College of Artificial Intelligence,Tianjin University of Science&Technology,Tianjin 300457,China;Food Safety Management and Strategy Research Center,Tianjin University of Science&Technology,Tianjin 300222,China)
出处
《计算机应用与软件》
北大核心
2021年第9期86-91,共6页
Computer Applications and Software
基金
天津市教委社会科学重大项目“大数据背景下食品安全网络舆情研究”(2017JWZD19)
天津市科技计划项目(17KPXMSF00140)。
关键词
个体相对权威性
HK模型
平均节点度
有限信任阈值
社交网络
Individual relative authority
HK model
Average degree of node
Bounded confidence threshold
Social network
作者简介
马永军,教授,主研领域:智能信息处理,食品安全;柴梦瑶,硕士生。
