在线社交网络的竞争性非线性动力学信息传播模型研究

Research on of Competitive Nonlinear Dynamic Information Diffusion Modeling in Online Social Network

针对在在线社交网络上如何用一种信息去抑制另一种信息的扩散传播,从而为商品营销和舆情管控提供理论支持的问题,本文提出了一种在线社交网络竞争性非线性动力学信息传播模型.首先,运用马尔可夫链理论分析网络上不同类型信息之间的竞争机制与节点状态转化、信息传播演化规律之间的内在联系,分别从概率视角构建了节点状态转化的概率模型,从统计视角构建了网络系统信息扩散的非线性动力学传播模型;然后,对提出模型的传播动力学微分方程求解平衡点并进行稳定性分析;最后,通过仿真分析模型中参数之间的相对变化关系、模拟了信息竞争和传播过程,并通过真实数据实证对比实验分析了模型仿真结果与真实事件统计数据的吻合程度,相似度和相关系数分别达到0.84和0.82以上.研究结果表明:提出的竞争性信息传播模型是合理、有效的,为解决在线社交网络上不同类型信息的竞争性传播类问题提供了一种新的科学方法和研究途径,具有较高的理论价值和应用价值.

In order to solve the problem of how to use one kind of information to restrain the diffusion and dissemination of another kind of information in online social networks,so as to provide theoretical support for commodity marketing and public opinion control,this paper proposes a competitive non-linear dynamic information dissemination model for online social networks.Firstly,competitive information propagation and mathematical model in online social network are summarized from three levels,such as micro level,medium level and macro level,and small-world network,scale-free network and online social network information popularity are defined respectively.The process of competitive information propagation and competitive behavior in online social network are described in this paper.The Markov chain theory is used to analyze the intrinsic relationship between the competition mechanism among different types of information on the network and the rules of node state transition and information propagation evolution.The macro probability model of node state transition is constructed from the perspective of probability,and the macro information diffusion of online social network system is constructed from the perspective of statistics.Then,we suppose that online social network is a closed network where information is generated and transmitted only in the social network,the total number of nodes on the network is stable.Combined with Routh stability determination principle,it gives the sufficient and necessary conditions to determine the stability of the system.When the basic regeneration number is less than 1,the equilibrium point is globally asymptotically stable.The equilibrium point and stability of the differential equation of the propagation dynamics of the model are solved to ensure that the model is mathematically reasonable and physically meaningful.Finally,the experiment aim and idea of it is clarified.The experimental procedure is given in detail in this paper.The feasibility analysis of the experimental results is carried out comprehensively.Five kinds of network structure diagrams are analyzed,which are chain structure,circle structure,star structure,group structure and complex network structure.Some hot real events in online social network are selected such as international event,science and technology event,business event,cultural event,education event,social event,the law rule event,these typical real emergency hot events are simulated and analyzed.It can verify the rationality of the proposed model from multiple perspectives by combining the Baidu index data.The relationship between the parameters in the proposed model is analyzed by numerical simulation experiment,and the dynamic simulation is carried out by simulation experiment.The process of information competition and dissemination is analyzed,and the degree of agreement between the simulation results and the real event statistics is analyzed through empirical comparative experiments.The similarity is above 0.84 and the correlation coefficient is above 0.82.A series of experimental results show that the proposed competitive information dissemination model in social network is reasonable and effective.It provides a new scientific method and research approach to solve the problem of competitive dissemination of different types of information on online social networks,and has high theoretical value and application value.