基于回音室超网络的网络舆情差异化引导研究

Research on Differentiated Guidance of Network Public Opinion Based on Echo Chamber Super-network

[目的/意义]针对网络舆情传播中的回音室现象,提出一种基于回音室超网络的网络舆情差异化引导模型,为网络舆情管控者制定引导方案提供实践支持。[方法/过程]首先,文章构建了用户、事件、回音室三个子网络,并根据用户间复杂交互行为生成三个子网络间的映射关系,进而构建回音室超网络。其次,利用自然语言处理等技术对事件网络中的舆情信息进行特征向量提取及相似度计算,从而筛选出目标回音室,根据映射关系识别出目标用户。最后,建立基于回音室超网络的网络舆情差异化引导模型——根据舆情事件情感极性的不同对目标回音室用户采取不同引导策略。[结果/结论]采集了包含40条舆情事件、26943条转发评论内容和26113名用户的训练数据进行试验分析,结果表明模型能较好地对目标用户实现有效引导从而影响舆情态势,为网络舆情的治理提供理论参考。

[Purpose/significance]In view of the echo chamber phenomenon in the dissemination of network public opinion,this study proposes a differentiated guidance model for network public opinion based on echo chamber super-network,which provides practical support for network public opinion controllers to formulate guidance schemes.[Method/process]Firstly,this paper constructs three sub-networks of users,events and echo chambers,and generates the mapping relationship between the three subnetworks according to the complex interaction behaviors among users,and then constructs the echo chamber supernetwork.Secondly,NLP and other technologies are used to extract feature vector and similarity calculation for the public opinion information in the event network,so as to screen out the target echo chamber and identify the target user according to the mapping relationship.Finally,a differentiated guidance model of network public opinion based on echo chamber super-network is established,which adopts different guidance strategies for target echo chamber users according to the different emotional polarities of public opinion events.[Result/conclusion]This paper collects data including 40 public opinion events,26943 forwarded comments and 26113 users for experimental analysis.The results show that the model can effectively guide the target users and thus affect the public opinion situation,and provide theoretical reference for the governance of network public opinion.