基于平台推荐机制和有界信任影响的网络群体观点演化研究  

Evolution of Network Group Views Based on Platform Recommendation Mechanisms and Bounded Trust Influences

作　　者：祁凯[1] 黎冰雪 杨志 Qi Kai;Li Bingxue;Yang Zhi(School of Management,Harbin Normal University,Harbin 150025;School of Economics and Management,Harbin Engineering University,Harbin 150001)

机构地区：[1]哈尔滨师范大学管理学院,哈尔滨150025 [2]哈尔滨工程大学经济管理学院,哈尔滨150001

出　　处：《情报杂志》2021年第5期132-142,185,共12页Journal of Intelligence

基　　金：黑龙江省哲学社会科学研究规划项目“大数据环境下突发公共危机事件风险识别及网络社会治理研究”(编号:17GLB025);黑龙江省博士后科研启动基金项目“突发公共危机事件网络集群动态演化规律及多元主体协同治理研究”(编号:LBH-Q19114)研究成果之一。

摘　　要：[目的/意义]基于现实背景扩展了Deffuant-Weisbuch(DW)模型,并对网络媒体平台中群体观点演化规律进行了系统研究,这对开发舆情预警系统以及完善舆情导控政策都具有一定的理论价值和社会实践意义。[方法/过程]在是否存在智能推荐机制的信息推送机制的基础上,考虑平台影响力、话题类型等因素,运用仿真实验分析了不同参数对群体观点演化过程的影响。[结果/结论]平台影响力和智能推荐机制都会影响网络群体观点的收敛速率;在无智能推荐机制情景下,对于敏感程度不同的话题,其网络群体观点分裂程度依旧具有显著差异。[Purpose/Significance]The Deffuant-Weisbuch(DW)model is extended based on the factual background,and the evolution of group views in online media platforms are systematically studied,which has absolute theoretical value and social practice significance for the development of public opinion early warning systems and the improvement of public opinion guidance and control policies.[Method/Process]Based on whether there is an intelligent recommendation mechanism for message pushing,the authors analyzed the influence of different parameters on the evolution of group views using simulation experiments,taking into account factors such as platform influence and topic type.[Result/Conclusion]Both platform influence and intelligent recommendation mechanism affect the convergence rate of group viewpoints;in the absence of an intelligent recommendation mechanism,the degree of splitting group viewpoints is still significantly different for topics with different sensitivity levels.

关 键 词：DW模型 群体观点演化 网络舆情 推荐机制 信息茧房 

分 类 号：C931.1[经济管理—管理学]