融入Shapley值合作博弈的社交网络用户舆情控制模型研究
作者:
作者单位:

浙江工商大学

作者简介:

通讯作者:

中图分类号:

C912.63

基金项目:

国家自然科学基金项目(面上项目,重点项目,重大项目)


Research on Social Network User Public Opinion Control Model Integrated with Shapley Value Cooperative Game
Author:
Affiliation:

Zhejiang Gongshang University

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    本文旨在寻找最优节点来对社交网络中的舆情信息进行控制,同时结合用户群体中心性定义社交网络舆情控制准则,以识别社交网络中舆情信息传播的关键节点。此外,本文还提出了一种基于用户群体的合作博弈收益分配方法,利用Shapley值分配不同用户群体中用户通过相互进行位置权力和功能所获得的盈余,将识别出的关键节点标记为基于Shapley值的信息分隔符(SVID),并提出一种自适应的高效SVIDA算法。最后结合SIR模型和真实网络数据集进行数值仿真和算法比较,以验证了本文所提方法的有效性。实验结果显示:本文所提SVIDA法可以在同质、异质和国内外真实社交网络环境中进行舆情控制,且效率较其他舆情控制算法有较大提升;就SIR模型仿真结果可知,当分隔符比例q提升时,其受感染节点的比例下降趋势加速,且免疫节点比例上升的速度加快;在传播速度更快的网络环境中,本文所提SVIDA法的舆情控制优势更明显。

    Abstract:

    This paper aims to find the optimal node to control the public opinion information in the social network, and define the public opinion control criteria of the social network in combination with the centrality of the user group, so as to identify the key nodes of the dissemination of public opinion information in the social network. In addition, this paper also proposes a cooperative game income allocation method based on user groups, using Shapley value distribution of different user groups of users through the mutual positions of power and function of surplus, to identify the key node is marked as separators (SVID) based on the Shapley value of information, and puts forward an adaptive efficient SVIDA algorithm. Finally, SIR model and real network data sets are combined to carry out numerical simulation and algorithm comparison to verify the effectiveness of the proposed method. The experimental results show that the SVIDA method proposed in this paper can control public opinion in homogeneous, heterogeneous and real social network environment at home and abroad, and its efficiency is greatly improved compared with other public opinion control algorithms. According to the SIR model simulation results, when the proportion of delimiter q is increased, the proportion of infected nodes decreases faster, and the proportion of immune nodes increases faster. In the network environment with faster transmission speed, the public opinion control advantages of SVIDA method proposed in this paper are more obvious.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-03-19
  • 最后修改日期:2020-06-04
  • 录用日期:2020-06-12
  • 在线发布日期:
  • 出版日期: