Research On Network Opinion Distortion Model Based On Elite User Silence

The rise of social networking platforms has provided new ways and media for the development and evolution of public opinion,the process of information dissemination has become more rapid and diversified,and more and more users have participated in the discussion of social topics.It is in such a new media context with complex information dissemination process and uneven network users that the news reports in the early stage of the outbreak on hot social topics often distort the truth of the matter,and there is a big deviation between the early reports and the follow-up reports,forming a phenomenon of network opinion distortion.This kind of phenomenon often leads to the network public opinion environment difficult to control,and cause bad social influence,so it is of great practical significance to study the causes,key factors and development rules of network opinion distortion to effectively govern the network public opinion environment.On the basis of sorting out relevant research on Internet opinions at home and abroad,this paper improves the traditional HK model by combining with the bounded trust principle,optimizing various parameters in the traditional model,integrating elite users and super opinion leaders and using Matlab simulation software to simulation the four models,so as to obtain the law of network opinion distortion phenomenon,and give some suggestions from three aspects about the government,the media and the audience to avoid network opinion distorted and optimize the environment of network public.The conclusion of this paper is mainly drawn from the following two aspects of research:On the one hand,a network opinion evolution model based on elite users is designed.Based on the analysis and summary of the existing role-based research on network public opinion communication,this paper innovatively proposes the role of elite user,and summarizes the differences between elite users and super opinion leaders,and their own characteristics.That is,the number of super opinion leaders in the network is very small,and they will not be influenced by the surrounding users and not easily change the value of their opinions.They have extremely high authority and influence in the network,and the dissemination of opinions is unlimited;Elite users also have a higher level of authority andtrust in the network,but their influence is lower than that of super opinion leaders,and they will change their opinion value under the influence of the opinions of their neighbors.Considering that different individuals have different influence coefficients,different levels of trust between different individuals and different distances in the network,this paper optimizes the weight of influence and the measurement of trust threshold in the traditional model,and establishes the elite user influence model based on the improved HK model to study the change of the overall viewpoint trend of the network under the influence of elite users when there are no super opinion leaders in the network,and find that elite users can converge the opinions of most nodes in the network to those of elite users,but cannot change some extreme nodes that hold opposite opinions.Then,add an super opinion leader node that contradicts the views of elite users in the network and establish the network viewpoint evolution model that integrates an super opinion leader,and the simulation experiment is carried out to observe the influence of super opinion leaders on network opinion distortion,and find the "elite user compromise point" and "elite user convergence point" these two key stage division points.This part of the model and the corresponding simulation results lay a foundation for the following elite user silence model.On the other hand,aiming at the uncertainty of user opinions in environmental noise in the network and combining the advantages of gaussian white noise in quantitative modeling and derivation analysis,an elite user silence model based on super opinion leaders was established.Firstly,a silence model of elite users under the influence of single super opinion leader is proposed,considering the influence of elite users on the overall evolution of the network by reducing the frequency of public opinions under group pressure caused by environmental noise when there is only one super opinion leader node that is contrary to the views of elite users in the network,that is,the influence of the lack of exchange and contrast opinions of elite users on the overall evolution of the network.Then,in order to break the "silent spiral" phenomenon formed by group pressure,the paper adds a new super opinion leader node of elite user support party on the basis of the single super opinion leader model,and proposes the elite user silence model under the competition of double super opinion leaders,exploring the improvement of online opinion distortion by elite users when they are supported by authorities.At the same time,the simulation results of the model can also inspirethe effective governance of network public opinion environment.