Journal of Computer Applications ›› 2017, Vol. 37 ›› Issue (5): 1419-1423.DOI: 10.11772/j.issn.1001-9081.2017.05.1419

Previous Articles     Next Articles

Trend prediction of public opinion propagation based on parameter inversion — an empirical study on Sina micro-blog

LIU Qiaoling, LI Jin, XIAO Renbin   

  1. School of Automation, Huazhong University of Science and Technology, Wuhan Hubei 430074, China
  • Received:2016-11-14 Revised:2016-12-14 Online:2017-05-10 Published:2017-05-16
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61540032).

基于参数反演的网络舆情传播趋势预测——以新浪微博为例

刘巧玲, 李劲, 肖人彬   

  1. 华中科技大学 自动化学院, 武汉 430074
  • 通讯作者: 李劲
  • 作者简介:刘巧玲(1993-),女,湖北当阳人,硕士研究生,主要研究方向:舆情传播、大数据;李劲(1980-),男,湖北武汉人,博士,主要研究方向:舆情大数据、神经网络;肖人彬(1965-),男,湖北武汉人,教授,博士,主要研究方向:复杂系统、复杂社会管理。
  • 基金资助:
    国家自然科学基金资助项目(61540032)。

Abstract: Concerning that the existing researches on public opinion propagation model are seldom combined with the practical opinion data and digging out the inherent law of public opinion propagation from the opinion big data is becoming an urgent problem, a parameter inversion algorithm of public opinion propagation model using neural network was proposed based on the practical opinion big data. A network opinion propagation model was constructed by improving the classical disease spreading Susceptible-Infective-Recovered (SIR) model. Based on this model, the parameter inversion algorithm was used to predict the network public opinion's trend of actual cases. The proposed algorithm could accurately predict the specific heat value of public opinion compared with Markov prediction model.The experimental results show that the proposed algorithm has certain superiority in prediction and can be used for data fitting, process simulation and trend prediction of network emergency spreading.

Key words: Sina micro-blog, Susceptible-Infective-Recovered (SIR) model, Back-Propagation (BP) neural network, parameter inversion, public opinion propagation

摘要: 针对现有的舆情传播模型研究与实际舆情数据结合较少以及难以从舆情大数据中挖掘舆情传播内在规律的问题,提出一种基于实际网络舆情大数据采用神经网络的舆情传播模型参数反演算法。改进经典SIR传染病传播模型,构建一种网络舆情传播模型,基于该模型对实际案例进行参数反演,预测网络舆情的后续传播趋势,并与马尔可夫预测模型对比,所提算法可以精确预测舆情的具体热度值。实验结果表明,所提算法在预测性能上具有一定的优越性,可以用于网络突发事件传播的数据拟合、过程模拟和趋势预测。

关键词: 新浪微博, SIR模型, 反向传播神经网络, 参数反演, 舆情传播

CLC Number: