复杂网络中谣言传播模型及防控策略研究

发布时间：2018-01-22 08:56

本文关键词： 谣言传播社交网络平均场方程传播阈值辟谣策略　出处：《上海大学》2016年博士论文　论文类型：学位论文

【摘要】：社交网络平台的发展和普及使谣言制造和传播更加便利,严重威胁着社会的安定。本文主要研究了复杂网络上的谣言传播规律,并且探讨了谣言传播的应对策略。1.研究了非均匀网络中带有遗忘和记忆机制的谣言传播机理,发现了网络结构对谣言传播的影响。将带有遗忘和记忆机制的谣言传播模型引入到非均匀网络中,并建立了平均场方程,通过稳态分析得出了此谣言传播模型在非均匀网络中不存在传播阈值的结论。通过数值仿真发现：网络的拓扑结构会影响谣言的传播过程,相比于均匀网络,在非均匀网络上谣言传播更快、结束更晚但最终规模更小；遗忘机制和记忆机制对谣言的最终规模也有影响,遗忘率越大,谣言的最终规模越小；记忆率越大,谣言的最终规模越大。2.建立了考虑传播参数随时间变化的谣言传播模型,并与恒定传播参数的谣言传播模型进行了比较。政府的干预及人类自身的遗忘特点使谣言传播参数(传播率、停止率和遗忘率)具有随时间变化的特性,将这些可变的传播参数引入经典的谣言传播模型中,建立了平均场方程,并且通过稳态分析研究了传播阈值。通过数值仿真发现：网络结构层次越高,谣言传播越快、持续时间越长、最终规模却越小；相比恒定参数的谣言传播,变化的传播率和遗忘率都使谣言传播更快且更广,变化的停止率使谣言传播更慢却更广。3.建立了带有真相传播者的谣言传播模型,分析了真相传播者对谣言传播的抑制作用。考虑到真相传播者能起到辟谣作用,构建了带有这种辟谣机制的谣言传播模型,建立了平均场方程,并进行了稳定性分析。通过数值仿真发现：真相传播者抑制谣言传播速度和规模从而降低了谣言影响力；一般情况下,规则网络中的谣言最终规模大于BA无标度网络中规模,但在真相传播率较高的条件下,规则网络中的谣言最终规模小于BA无标度网络中规模。4.建立了考虑不同辟谣方式的谣言传播模型,分析了这些辟谣方式对谣言传播的影响。构建了带有专业辟谣账号、举报以及系统通知三种辟谣方式的谣言传播模型,建立了平均场方程,进行了稳定性分析,通过数值仿真发现：专业辟谣账号加强普及科学真相所产生的抑制谣言传播的效果要好于加强说服谣言传播者停止传播的效果；对谣言有辨别力的用户对谣言的举报率、参与辟谣的积极性以及所占人群的比例都对辟谣效果有影响；越早发布系统通知都使得辟谣效果越好。
[Abstract]:The development and popularization of social network platform make rumor making and spreading more convenient and threaten the stability of society seriously. This paper mainly studies the law of rumor spread on complex network. And discussed the rumor communication strategy. 1. Studied the non-uniform network with forgetting and memory mechanism of rumor communication mechanism. 2. The influence of network structure on rumor propagation is found. The rumor propagation model with forgetting and memory mechanism is introduced into the non-uniform network and the mean field equation is established. Through steady-state analysis, we get the conclusion that there is no propagation threshold in the non-uniform network. Through numerical simulation, we find that the topology of the network will affect the rumor propagation process, compared with the uniform network. Rumors spread faster on uneven networks, ending later but eventually on a smaller scale; The forgetting mechanism and memory mechanism also have influence on the final scale of rumors, the larger the forgetting rate, the smaller the final scale of rumors; The larger the memory rate, the larger the final scale of rumors. 2. A rumour propagation model considering the variation of propagation parameters over time is established. And compared with the constant spread parameters of rumor propagation model. The government intervention and human own forgetting characteristics make the rumor propagation parameters (propagation rate, stop rate and forgetting rate) have the characteristics of changing with time. These variable propagation parameters are introduced into the classical rumor propagation model, the mean field equation is established, and the propagation threshold is studied by steady-state analysis. The numerical simulation shows that the higher the network level is, the higher the network structure is. The faster the rumors spread, the longer the duration, but the smaller the final scale; Compared with constant parameter rumor propagation, both the changing propagation rate and the forgetting rate make the rumor spread faster and wider. The changing stoppage rate makes the rumor spread slower but wider. 3.The rumor propagation model with truth communicator is established. This paper analyzes the inhibitory effect of truth communicators on the spread of rumors. Considering that truth communicators can play a role in refuting rumors, a rumor propagation model with this mechanism is constructed, and the mean field equation is established. Through the numerical simulation, it is found that truth communicators suppress the speed and scale of rumor propagation and thus reduce the influence of rumors; In general, the scale of rumors in regular network is larger than that in BA scale-free network, but under the condition of high truth propagation rate. The final scale of rumour in regular network is smaller than that in BA scale-free network. 4. A rumour propagation model considering different ways of refuting rumor is established. This paper analyzes the influence of these disinformation refuting methods on rumor propagation. A rumor propagation model with professional rumour-refuting account, reporting and systematic notification is constructed, and the mean field equation is established, and the stability analysis is carried out. Through numerical simulation, it is found that the effect of spreading rumors by strengthening the popularization of scientific truth is better than that of persuading rumor disseminators to stop spreading. The report rate of the users who have the discriminating power to the rumor, the enthusiasm of participating in refuting the rumor and the proportion of the crowd all have influence on the rumour refuting effect; The early release of system notifications will make the rumour disclaimer the better.
【学位授予单位】：上海大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：G206

【相似文献】