Research On Rumor Spreading Model And Control Strategy Considering Media Intervention

With the advent of the Web2.0 era,people can express their thoughts and spread the news on the internet at any time.This provides a more convenient way to spread the rumors,so that the scope of the rumor spreading has been expanded and the propagation speed has been faster,which means it's easier to affect people's daily work and life.Therefore,the construction of a more realistic rumor propagation model is helpful to better understand the spreading mechanism of rumors.Study on the control strategies of rumor propagation can help to reduce the harm to society caused by the rumor.The main work of this paper includes the following respects:1.Sina micro-blog has a huge amount of registered users and large amount of information interaction,which is a representative social platform.The user relationship data is collected on sina micro-blog through web crawler.The data is analyzed and converted into adjacency matrix to construct the user relationship network diagram.The degree distribution,clustering coefficient and other characteristic parameters of the network are analyzed and compared with the other network topology generated by model algorithms.The result shows that the data collected can reflect the real social network characteristics better,which can be used as the base map to run the rumor spreading model.In addition,we write the web crawler to crawl a micro-blog rumor about haze,and analyzed its actual spreading process,which providing the basis for the following research of rumor spreading model.2.Considering the fact that not all users in the network will choose to spread the rumor when they are exposed it at the first time.Besides,the impact of rumor spreading caused by the media clarification should also be considered.So,by improving the existing rumor propagation model,a new rumor spreading model under the media intervention is constructed.The numerical analysis results show that the credibility of media has little effect on the final scale of rumor propagation,and the simulation results verify the conclusion.In addition,the influence of media clarification time and the credibility of media on the evolution of each node in the process of rumor propagation is analyzed.The greater the credibility of the media,the faster the spread node attenuation rate and the faster the immune node rise rate,which can accelerate the rumor of the decline and reducing the rumors of the life cycle to some extent.3.Two rumor control strategies are proposed.The first is local control strategy.By selecting a certain percentage of nodes randomly in the network,and finding out the propagation nodes among them at the time of applying control.Then,to control the vulnerable nodes within a certain steps around those propagation nodes,the final rumor spreading scale can be reduced.Besides,the performance index is defined according to the proportion of control nodes in the network,to find the optimal control step.The second one is improved acquaintance control strategy,which is based on the acquaintance immunization strategy.However,the control target nodes selection principle has been optimized from the network structure and the node behavior.Based on the above rumor propagation model,simulation experiments were carried out both before and after control,and compared with other immune strategies.The results show that both local control and improved acquaintance control can reduce the scale of rumor propagation in social networks to some extent,and are effective control strategy.At last,we analyzed the applicability of the two strategies.