融入心理因素的在线社交网络谣言传播模型

An Online Social Network Rumor Propagation Model Integrated with Psychological Factors

随着社交平台的兴起,谣言会在短时间内快速扩散,对社会造成负面影响,因此研究谣言传播过程中的影响因素迫在眉睫,这对政府把控和引导突发事件的舆论起着关键性作用。而目前考虑心理因素对谣言传播过程影响的研究较少,因此该模型在经典传染病模型SEIR模型的基础上,将总人群分为四类:易感状态节点、潜伏状态节点、感染状态节点和免疫状态节点。考虑好奇心心理、从众心理、信息干扰率和个体免疫率四种因素,构建融入心理因素的改进SEIR谣言传播模型。首先,利用微分动力学方法和下一代矩阵理论计算了该模型的平衡点,分析并证明了平衡点的局部渐近稳定性。接着使用MATLAB对模型在自建网络和真实网络中进行仿真实验,数值仿真结果表明:好奇心越强烈,由易感状态节点转变为潜伏状态节点的速度越快,潜伏状态节点的峰值越大;从众心理越强烈,由潜伏状态节点转变为感染状态节点的速度越快,感染状态节点的峰值也越大。适当提高个体免疫率和信息干扰率可以缩短谣言传播时间,降低谣言传播总量。

With the rise of social platforms,rumors will spread rapidly in a short period of time and have a negative impact on society.Therefore,it is urgent to study the influencing factors in the process of rumors propagation,which plays a key role in the government's control and guidance of public opinion in emergencies.At present,there are few studies that consider the influence of psychological factors on the spread of rumors.Therefore,this model divides the total population into four categories based on the classic infectious disease model SEIR model:susceptible state nodes,exposed state nodes,infected state nodes and removed status node.Considering four factors:curiosity,herd mentality,information interference rate and individual immunity rate,we construct an improved SEIR rumor propagation model that incorporates psychological factors.First,the equilibrium point of the model is calculated using the differential dynamics method and the next-generation matrix theory,and the local asymptotic stability of the equilibrium point is analyzed and proved.Then MATLAB software is used to perform simulation experiments on the model in the synthetic network and real network.The numerical simulation results show that the stronger the curiosity,the faster the transition from the susceptible state node to the exposed state node,and the greater the peak value of the exposed state node;the stronger the herd mentality,the faster the transition from the exposed state node to the infected state node,and the greater the peak of the infected state node.Appropriately increasing the rate of individual immunity and the rate of information interference can shorten the spread of rumors and reduce the total amount of rumors spread.