基于信息传播模型-SIR传染病模型的社交网络舆情传播动力学模型研究

发布时间：2018-06-05 06:54

本文选题：SIR模型 + 舆情传播　；参考：《情报科学》2017年12期

【摘要】：【目的/意义】随着移动网络技术的飞速发展,用户已习惯在社交网络平台发表意见,进而形成所谓的网络舆情。准确建立舆情的传播模型,对于舆情的引导和控制具有重要的帮助。【方法/过程】本文基于传统的SIR传染病模型,综合考虑用户的心理特征行为因素,搭建新型的社交网络舆情传播动力学模型,并选用粒子群算法,以2016年内热点的微博舆情事件为例,求解模型参数的最优值,并进行实验数据验证。【结果/结论】结果表明:用户的追根溯源心理、持续关注心理以及漠不关心心理等心理特征对舆情的传播特性有重要影响,同时本文给出的模型由于考虑了用户的心理特征行为因素,模型的准确性相较传统的SIR模型有明显优势,模型拟合曲线与真实数据曲线基本一致,并且模型拟合值与真实数据的绝对误差值和RMSE值都较低。本文的研究对准确预测舆情信息传播趋势以及舆情的分析和引导有着重要的指导作用。
[Abstract]:With the rapid development of mobile network technology, users have become accustomed to express their opinions on social network platform, and then form the so-called network public opinion. It is of great help to establish an accurate communication model of public opinion for the guidance and control of public opinion. [method / process] based on the traditional SIR infectious disease model, this paper comprehensively considers the psychological characteristics and behavioral factors of users. A new dynamic model of public opinion propagation in social networks is built, and the particle swarm optimization algorithm is used to solve the optimal parameters of the model by taking the Weibo public opinion events of hot spots in 2016 as an example. The results show that the psychological characteristics of the users, such as tracing back to the source, constant concern and indifference, have an important effect on the communication characteristics of public opinion. At the same time, the model given in this paper has obvious advantages over the traditional SIR model because of taking into account the factors of user's psychological characteristics and behavior, and the fitting curve of the model is basically consistent with the real data curve. And the absolute error value and RMSE value of model fitting value and real data are lower. The research in this paper plays an important guiding role in accurately predicting the trend of public opinion information dissemination and the analysis and guidance of public opinion.
【作者单位】：江西师范大学传播学院;
【基金】：国家社科基金项目(13CXW001)
【分类号】：G206;R181

【相似文献】