Studies On Information Spreading In Social Network

As communication technology advances,information dissemination becomes more complex and diverse,and the traditional information spreading model can no longer truly reflect the process of information spreading in social networks.At present,information spreading research based on social network has attracted the interest of researchers at home and abroad.Based on previous studies,we explore how different information intervention factors,network structure and individual heterogeneity affect the evolutional process of information.This paper mainly includes the following two aspects:(1)Influence of information attraction and forgetting factors on information spreading in different static networks are studied.Considering the attraction of information and individual forgetting factors,we propose an susceptible-exposed-infected-recovered(SEIR)information spreading model.Firstly,we obtain the theoretical spreading threshold on homogeneous and heterogeneous networks by using the mean-field method,and its accuracy is verified.Secondly,numerical experiments show that compared with ER random network,the spreading threshold on BA scale-free network is smaller and the propagation speed is faster.Interestingly,information attraction will have different effects on the process of information spreading in ER network and BA network.In BA network,information attraction will significantly promote information diffusion,while in ER network,information attraction will only have an impact in a lower range.Finally,the results show that forgetting rate can promotes information spreading in two different network structures,and more obvious in BA network.(2)Information spreading in two-layer time-varying activity-driven network are studied.Considering the influence of different information spreading channels,time-varying characteristics of network structure and individual heterogeneity on information spreading,a novel two-layer time-varying information spreading model based on activity-driven network is proposed.The paper obtain theoretical spreading threshold by microscopic Markov chain approach,and simulate a large number of numerical simulations.First of all,compared with offline spreading channels,online spreading channels have a more significant impact on information spreading.Secondly,when individual activity rate and spreading ability obey power law distribution respectively,heterogeneity will promote information spreading.However,when individual spreading ability obeys Gaussian distribution,the change of heterogeneity has little effect on information spreading.Finally,in a certain range,the increase of individual contact capacity will promote the information spreading,but when it reaches a certain value,its further increase will not have an impact on the information spreading process.