Modeling And Simulation Of Dynamic Emotional Entrainment In Public Agendas

The sudden outbreak of COVID-19 pandemic in 2020 has plunged countries around the world into varying degrees of chaos.Since face-to-face contact is severely restricted,online platform has functioned as an important alternative channel beyond distance.Emotional entrainment has become more intense and frequent during this period.On the one hand,the pandemic of infectious diseases announces that the entire society has entered a state of emergency.The COVID-19 virus can quickly infect through the respiratory tract and other channels,turning uninfected people into infected ones.The ”pervasive invisible killer” creates a special social environment for the spread of panic;On the other hand,the pandemic of infectious diseases has severely reduced the living standards of some peo-ple,leading to the intensification of inherent social contradictions.Potential negative social emotions arouse in daily discussions of public issues.In the end,the emotional flow in the social communication system drowned out the information flow,exposed more frequently in the field of public opinion.Based on the background of post-pandemic era,this thesis studies the general laws of emotional entrainment on social networks.The research is not constrained by the perspective of emotional dif-fusion in emergencies,but fully considers both subjective and objective variables in social communi-cation system that may inluence normal emotional entrainment.Therefore,the study is of universal significance and can provide beneficial enlightenment for public opinion monitoring and guidance.In terms of research methods,this thesis applies modeling and simulation technology in the field of computer science,and proposes a special dynamic emotional entrainment model.This model simulates the emotional interaction rules in the real world based on the theory of opinion dynamics.At the same time,dimensional emotion model proposed by psychologists is applied to vectorized emotional entrainment.In addition,based on data mining and automated information processing and analysis,the paper obtains the relevant information of Twitter users and constructs a real social network.The effectiveness of the model is verified through comparative experiments.The experimental simulation results show that: in a social network with small-world characteristics1.If there are a lot of extreme emotions in initial emotions,it will take longer time for public emotions to interact and finally achieve convergence.However,some extreme emotions can still exist.2.In a society that people fully trust each other,even if people with high emotional valence and arousal and low emotional valence and arousal exist at the same time,eventually the public emotions can synchronize to a moderate state.The same emotional entrainment results may also appear in intimate groups.3.When a node with high emotional valence highly trusts its neighbors and has a lower trust threshold in a certain public agenda,the overall social emotions are more likely to converge.4.The higher an individual persists to its own emotions,the longer it will take for the overall social emotions to converge.The simulation results provide the following enlightenment for public opinion monitoring and guidance: 1.Pay attention to public agendas that may arouse more polarized initial emotions.2.Build a trustworthy social environment and pay attention to issues that involve a majority of stakeholders.3.The public opinion guidance needs to go deep into marginal groups where may exist polarized emo-tions.4.Show more patience with opinion guidance in agendas that involve a lot of stubborn groups and pay attention to groups with high stickiness.5.Publicize malicious instigators and reduce public trust in them.