网络舆情的突变性扩散方向估计建模仿真  

Mutation Diffusion Direction Estimation Modeling and Simulation for Network Public Opinion

作　　者：何英[1] 胡之骐[2] 

机构地区：[1]四川外国语大学教育技术中心,重庆400031 [2]重庆师范大学教育科学学院,重庆401331

出　　处：《计算机仿真》2017年第3期281-284,360,共5页Computer Simulation

摘　　要：对网络舆情的突变性扩散方向进行准确估计,可以提高政府对网络突发事件的应对能力。需要利用网络舆情的时间关联度特征,对突变性扩散方向估计偏差进行补偿,建立扩散方向估计模型,但传统的尖点突变方法是利用尖点突变原理建立舆情传播模型,并建立突变性扩散方向估计模型,不能提取网络舆情的时间关联度特征,存在建模误差大的问题。提出一种改进灰色预测算法的网络舆情的突变性扩散方向估计建模方法。首先利用LDA模型提取网络舆情扩散时间关联度特征,组建网络舆情的突变性扩散文本集,得到网络舆情突变性扩散的时间热度值,建立基于灰色预测的网络舆情的突变性扩散方向估计模型,并利用嵌套法对模型进行求解。仿真结果表明,所提算法对网络舆情的突变性扩散方向估计精确度较高。Accurate estimation of the abrupt diffusion direction of network public opinion can improve the ability of the government to deal with the network emergencies.It needs to use the time correlation characteristics of network public opinion,to compensate for the deviation of the direction of the sudden change diffusion,and to establish the model of diffusion direction estimation.But traditional cusp mutation method makes use of the cusp mutation principle to establish the public opinion propagation model,and establishes the model of the diffusion direction estimation of the sudden change.It cannot extract the characteristics of the time correlation of network public opinion,and there is a big modeling errors.This paper proposes a modeling method for estimating the direction of abrupt diffusion based on network public opinion of improved grey prediction algorithm.Firstly,the LDA model is used to extract diffusion time correlation characteristics of network public opinion.The abrupt diffusion text set of network public opinion is set up to get the time heat value of abrupt diffusion of network public opinion.And then,a model for estimating the abrupt diffusion direction of network public opinion is established based on grey prediction,and the nesting method is used to solve the model.The simulation results show that the proposed algorithm has a high accuracy in the estimation of the diffusion direction of the network public opinion.

关 键 词：网络舆情 突变性扩散方向 估计模型 

分 类 号：G203[文化科学—传播学]