基于D-S证据理论的时序网络节点重要性研究
Research on Node Importance Identification for Temporal Network Based on D-S Evidence Theory
张李林清 1张琨 1吕来水1
作者信息
- 1. 南京理工大学计算机科学与技术学院 南京 210000
- 折叠
摘要
特征向量中心性是衡量复杂网络节点重要性的有效方法之一,但时序网络特征向量中心性无法衡量节点的全局重要性.因此引入D-S证据理论,将每一时间层的节点重要性作为一个信息源,通过Dempster规则进行多源信息融合从而得到节点的全局重要性.在Enron、Workspace实证网络数据集上的实验结果显示,基于D-S证据理论的时序网络特征向量中心性比其他方法的Spearman相关系数平均提高22.9%和21.8%,表明该方法能够有效地将时序网络特征向量中心性运用于衡量节点的全局重要性.
Abstract
The eigenvector centrality is one of the effective methods to measure the importance of complex network nodes,but the eigenvector centrality cannot measure the global importance of the nodes in temporal network.Therefore,D-S evidence theory is introduced.The importance of nodes in each time layer is taken as an information source,and multi-source information fusion is car-ried out by Dempster's rule to obtain the global importance of nodes.Experimental results on Enron and Workspace datasets show that the centrality of temporal network based on D-S evidence theory is 22.9%and 21.8%higher than the Spearman correlation coef-ficient of other methods.It is shown that the method can effectively apply the eigenvector centrality to measure the global importance of the nodes.
关键词
证据理论/特征向量中心性/节点重要性/时序网络Key words
evidence theory/eigenvector centrality/importance of nodes/temporal network引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据