节点文献

因特网上舆情传播的预测建模和仿真研究

Predictive Modeling & Simulation for Propagation of Internet Public Opinion

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 方薇何留进宋良图

【Author】 FANG Wei1,3 HE Liu-jin2 SONG Liang-tu1 (Institute of Intelligent Machines,Chinese Academy of Sciences,Hefei 230031,China)1(The HuaiNan Office of Information Application,Huainan 232000,China)2(College of Computer Science and Technology,University of Science and Technology of China,Hefei 230027,China)3

【机构】 中国科学院合肥智能机械研究所中国科学技术大学计算机科学与技术学院安徽省淮南市人民政府信息化工作办公室

【摘要】 网络舆情一定程度上表达出社会公众意愿,它虽然具有一般社会的舆论共性,但由于因特网的影响范围及传播速度,使其在虚拟社会中具有复杂系统的基本特征,故其传播倾向及发展方向受到重视。首先研究整体情感(正、负面)传播的预测模型及其算法;然后通过仿真找出影响其增长和消亡的规律。建模的出发点是将舆情传播看作一个时间序列的马尔科夫链;再利用哈肯协同理论提供的协同概率作为马尔科夫链的状态一步转移概率,构成一个协同-马尔科夫模型。在仿真实验中改变协同概率的各个参变量,以获取舆情随时间传播的不同演化过程的曲线簇,其结果可为虚拟社会管理提供参考。

【Abstract】 The Internet Public Opinion(IPO) usually expresses the voice of masses,although it hdue to the common attribute of the society opinion,but due to the large scope of influence and fast propagation on Internet,it has been attracted much attention to its tendency of propagation.The paper firstly studied the computer model and its algorithm to predict the propagation of IPO,i.e.the distributed probability of positive or negative opinion.Then the simulation gave the propagation order for growing or descending.In order to modeling,the propagation of IPO was observed as a random Markov time sequence chain,and its one step transition probability between states in the Markov chain was given by a synergistic probability which developeds Dr.H.Haken in his book of synergistic theory.A Synergistic-Markov model was proposed in the paper.By the simulation,the cluster of curves which expresses different evolving states of IPO propagation with time was depicted when we changed the variable parameters at the formula of synergistic probability.The result could provide reference for the management of virtual society.

【基金】 国家863计划(2006AA10Z23702);2009年度安徽省信息产业发展专项基金(财建[2009]722号)资助
  • 【文献出处】 计算机科学 ,Computer Science , 编辑部邮箱 ,2012年02期
  • 【分类号】TP393.09
  • 【被引频次】42
  • 【下载频次】923
节点文献中: