Research On The Factors And Its Influence Of The Evolution Of Public Opinion In Social Networks

The scale of Chinese netizens has exceeded 800 million now.The scale of social networks in China continue to expand rapidly because of the huge number of netizens.Social networks,such as Weibo and Wechat,have become the main platform for the spread of public opinion and influence.Public opinion in social networks can be regarded as the concentrated expression of real social public opinion to a certain extent.However,due to the opinions expressed by netizens are highly subjective and there is a lack of effective supervision,the evolution of network public opinion may develop in a negative direction,which will have a negative impact on social stability.Eliminating the negative influence of network public opinion and giving full play to the positive influence of network public opinion are of great significance to the country and society.It has also attracted a large number of scholars to conduct more in-depth research in the field of network public opinion.Therefore,this paper has important theoretical and practical significance for research on the factors and its influence of the evolution of public opinion in social networks.It is focusing on the factors of the evolution of network public opinion and its influence in this paper.We combined the characteristics of social networks,and studied from two aspects: the change of multi-information attention and the evolution of network public opinion.The formation and evolution of public opinion in social networks were analyzed in series,and a mathematical model which is in line with the actual situation of the network was proposed.The simulation experiments of the model were carried out to further explore the mechanism of the network public opinion.The contents and original contributions of this thesis are as follows:1.Multi-information concern coupling model was proposed.Whether information can attract enough attention of netizens decides whether public opinion can be formed,the period and influence of public opinion.Aiming at the problem of mutual coupling and attention of multi-information in real social networks,we proposed Multi-information concern coupling model.We were committed to analyzing the attributes of users and information,combining with the characteristics of most users' perception and attention to hot news information in the mainstream network information platform,to construct a network environment in line with the actual situation of human and information interaction,and enhance the applicability of the model.In order to improve the accuracy of the model and make the model more suitable for the actual situation,we creatively set up parameter mapping tables for some parameters in the model.These parameter mapping tables are based on the historical data of real news events,so it is still applicable to more types of news events.The establishment of parameter mapping table makes the model more scalable.The accuracy of our model is proved by comparing the simulation results with the actual data.Our research provides new ideas and methods for the research of multi-information attention.2.Group persuasion opinion evolution mode based on negative emotionalism was proposed.The opinions of different network groups directly constitute the network public opinions.For different types of network groups,we proposed group persuasion influence model based on negative emotionalism.We focus on the interaction process of individual and group opinions and the process of micro and macro mutual transformation.The rules of netizens' opinions interaction which are more in line with the characteristics of social network are redesigned.The new rule avoids the situation that the interaction process of individual opinions relies too much on the state of neighboring nodes and the network topology.We divided the influencing factors of individual decision-making into three aspects: the individual multiple attributes?the influence of network groups persuasion and the intervention of authoritative agencies.At the same time,we innovatively combined the heterogeneity of individuals with the theory of limited trust to put forward the individualized expression of the threshold of the difference degree of individual opinions.The mathematical expression avoids the appearance of threshold homogenization.Finally,we improved the cellular state change rules of cellular automata model.The improved cellular automata model is combined with the mathematical model proposed in this chapter and adopted by simulation experiments.Compared with the real news event data,the experimental results show that the error is acceptable,which verifies the relative accuracy of our model.Our research is of positive significance to further reveal the evolution mechanism of network public opinion.3.Experiments and results analysis were carried out for various influencing factors.We summarize some factors that may affect the netizens' attention and the evolution of network public opinion,including information type,content sensitivity,content relevance of update information,negative emotionalist groups,the intervention of authoritative agencies,synergistic factors and information receiving probability.Through the multi-information concern coupling model and group persuasion opinion evolution mode based on negative emotionalism which have been proposed in this paper as the mathematical basis,using MATLAB software as the simulation platform,the simulation experiments of each influencing factor are carried out one by one,and the experimental results are used as the basis to determine the influence of each factor,and then help us to achieve the ultimate goal of this study.