Research On Simulation Modeling Of Data Driven Public Opinion Information Dissemination On Social Networks

Recently,Internet technologies and online social networking services provide efficient and easy communication,enabling users to create,retrieve and disseminate messages on the go while making their messages widely available.Compared with mass media such as radio,television,and newspapers,online social networks have fundamentally changed the mechanism of information dissemination,and have gradually become the mainstream platform for information publication and dissemination.Research on social network information dissemination can not only help people understand the social network itself,explain the behaviour patterns of Internet users,and predict the trend of information diffusion,but also have wide application value in many fields such as public opinion regulation,precision marketing and information recommendation services.The characteristics of online social network information dissemination include massive data,diverse dissemination rules,spatio-temporal dynamics,non-linearity and so on.Taking social network public opinion as the application background,this paper studies the key technologies and methods of simulation modelling of data driven public opinion information dissemination on social networks.The main achievements and contributions of this paper are as follows:1.The network public opinion simulation framework based on parallel control,and the data-driven network public opinion simulation modelling framework are proposed.The data-driven network public opinion simulation modelling framework is the core component of the network public opinion simulation framework,including model-driven social network big data analysis,data-driven multi-disciplinary public opinion system modelling,and public opinion simulation system modelling.2.Based on the large-scale WeChat information cascading data,this paper studies the structure and spatio-temporal dynamics of social network information cascades,the pattern of social network information dissemination,and the influence of geographic factors,information contents on information dissemination.It is found that cascade sizes follow a power-law distribution,and the pattern of information dissemination is a mixture of broadcasting and viral spreading.It is found that the information dissemination probability follows a power-law distance decay effect,and both the probability and velocity of information dissemination show heterogeneity and diversity of geographic location.3.The RRT(Random Recursive Tree)is proposed to model the cascade growth process,and to describe the relationship between the average path length and the cascade size.Its single parameter quantifies the relative depth or width of cascade trees.The stochastic heterogeneous information dissemination model SVFR(Susceptible View Forward Removed)is proposed to depict the dynamic user behaviours including creating,viewing,forwarding and ignoring a message on a given social network.The SVFR model can explain the power-law cascade size distribution in We Chat,and can further describe the average path length and degree standard deviation of the cascade trees in relation to their sizes.4.The SVFR model is extended in time and space,and the models of CTSVFR(Continuous Time SVFR)and GIDDGNM(Geographic Interaction Data Driven Generative Network Model)are proposed.The CTSVFR model can characterize the information dissemination from temporal dimension.The GIDDGNM can reconstruct the modular and hierarchical structure,and geospatial characteristics of social networks based on partial observations of information cascade data,and have good results in multiple indicators such as degree distribution,average clustering coefficient,and average path length.In summary,combined with the advantages of data modelling and simulation modelling,this paper studies the key technology of simulation modelling of data-driven public opinion information dissemination in social networks.Based on the empirical data of We Chat information dissemination,the whole process from data mining analysis to the spatio-temporal modelling and simulation of information dissemination is validated.The research results are important for the social network public opinion regulation.