首页|融合用户传播风险和节点影响力分析的虚假信息传播控制方法

融合用户传播风险和节点影响力分析的虚假信息传播控制方法

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
在线社交网络中虚假信息传播蔓延成为当前网络空间安全治理面临的重要挑战.提出一种融合用户传播风险和节点影响力分析的虚假信息传播控制方法DDC-UPRNI(disinformation diffusion control method integ-rating user propagation risk and node influence analysis).综合考虑虚假信息传播特征空间的多样性和复杂性,通过自注意力机制实现用户传播虚假信息行为维度、时间维度和内容维度特征的嵌入表示,运用改进的无监督聚类K-means++算法实现不同用户传播风险等级的自动划分;设计一种自适应加权策略实现对离散粒子群优化算法的改进,进而提出一种基于离散粒子群优化的虚假信息传播关键节点选取方法,用于从具有特定传播风险等级的用户节点集合中选取若干个具有影响力的控制驱动节点,从而实现精准、高效的虚假信息传播控制;基于现实在线社交网络平台上开展试验,结果表明,所提出的DDC-UPRNI方法与现有算法相比,在控制效果和时间复杂度等重要指标上具有明显优势.该方法为社会网络空间中的虚假信息管控治理提供重要参考.
Disinformation diffusion control method integrating user propagation risk and node influence analysis
The spread of disinformation on online social networks(OSNs)has become a critical challenge for cyber-space security governance.This paper presents DDC-UPRNI,a disinformation diffusion control method,by integrating user propagation risk with node influence analysis.First,comprehensively considering the diversity and complexity of the characteristic space of disinformation propagation,an embedded representation of the behavior,time and content di-mensions of user propagation of disinformation is realized through the self-attention mechanism,and the automatic clas-sification of different user propagation risk levels is achieved using the improved unsupervised clustering K-means++ algorithm.Second,an adaptive weighting strategy is designed to improve the discrete particle swarm optimization al-gorithm,and a method for selecting key nodes of disinformation propagation is proposed based on the discrete particle swarm optimization.This method determines several influential control driving nodes from the user node set with a spe-cific propagation risk level to achieve accurate and highly efficient disinformation propagation control.Finally,experi-ments are performed on a real OSN platform,and the results demonstrate that the proposed DDC-UPRNI method has obvious advantages over other existing algorithms in some important indicators,including control effect and time com-plexity.This method provides a significant reference value for the current governance of disinformation in social cyber-space.

online social networksdisinformationpropagation riskembedded representationnode influenceadapt-ive weightingdiscrete particle swarmdiffusion control

荆军昌、张志勇、宋斌、班爱莹、高东钧

展开 >

河南科技大学 信息工程学院, 河南 洛阳 471023

河南科技大学 河南省网络空间安全应用国际联合实验室, 河南 洛阳 471023

在线社交网络 虚假信息 传播风险 嵌入表示 节点影响力 自适应加权 离散粒子群 传播控制

国家自然科学基金河南省中原科技创新领军人才项目中国博士后科学基金

619721332042005100212021M700885

2024

10.11992/tis.202210009

智能系统学报
中国人工智能学会　哈尔滨工程大学

智能系统学报

CSTPCD北大核心
影响因子：0.672
ISSN：1673-4785
年,卷(期)：2024.19(2)
  • 38