Modelling And Analysis Of Majority Opinion Based On Simplicial Opinion Space On Complex Social Networks

In recent years,with the development of science and technology,there are more and more different kinds of complex systems in contemporary life,such as social network,Internet and so on.Much information is transmitted and collided in these network structures.It play an important role in our lives.And social network is the most relevant to us.Therefore,the study of view integration base on social networking is practically significant for our life and formation of social opinion.The traditional classical model is mostly 0 and 1 mode,that is,agree or disagree.This paper uses concept of simplicial complex to describe view,each view contains multiple variables,which is closer to reality.This paper chooses a more realistic perspective,assuming that a view consists of multiple variables that together form a point of view.Then,the concept of simplex in the topology is used to describe the idea.The simplex in the topology can express the complex model in an intuitive way,and the Q-method is used to analyze the inherent properties and structure of the simplex.At the same time,the theory of infectious dynamics is extended to the dissemination of social network information.On the basis of these,this paper constructs the model of viewpoint collision,and then composes the MATLAB code to simulate the process of collision fusion.This paper explores the situation of collision fusion under different network structures.For the network structure,this paper chooses three classic network structures,namely,complete network,random network and scale-free network.The complete network is the simplest network structure,the scale-free network is closest to reality,and the random network is between the two.So choosing these three networks for simulation experiments,the results are more reliable and objective.From the simulation results,we found that in different network structures in the concept of fusion simulation curve,the overall trend of the final are flat,the trend is similar.Although the network structure will have some impact,but the basic trend of the same,and no decisive impact.But some specific parameters and specific network indicators have a huge impact on the results.In the simulation,for the same network structure,this paper uses different initial nodes,that is,the size of the network;different number of events,that is,how many different concepts and other parameters as the initial value,to simulate and compare their impact on the results.We found that the size of the network does not affect the overall trend of the fitting curve,but the larger the scale,the slower the rate of curve changes.At the same time,the more the number of events,the smaller the initial result of the fitting curve.And this paper explores the convergence of the same network in different topologies.In this paper,the network topology is simulated by the way of network failure,which is very consistent with the real situation.The real network,there will always be some node or side of the fault.In the topological network,the Beatty number is an important structural index,the 1? represents the connectivity,the 0? represents the number of connected branches.In this paper,different Betty numbers are selected to simulate the simulation experiment,and then compare the results of the fusion.From this article found and summed up some of the law,the greater the1?,the better the concept of integration,on the contrary,the integration is bad.