摘要
在线社交网络中关键节点的识别对研究网络的生存性和鲁棒性具有重要意义,现有众多关键节点研究大多从节点的局部环境及对网络功能的影响出发,并未考虑网络的全局拓扑结构.持续同调是计算拓扑的工具,可以捕捉高维的拓扑特征而被广泛应用于复杂网络的分析.文章首先定义了基于持续同调论的节点单形中心性,给出了新的在线社交网络中节点重要度描述指标.其次,提出了基于单形中心性的关键节点发现算法(KDSC)以得到社交网络中节点的重要度排序,从而系统地给出了一种基于持续同调的关键节点度量和发现方法.在验证实验中,文章对真实社交网络的节点单形中心性进行评价和分析,与传统度量指标进行对比讨论;利用KDSC求解关键节点,并与经典关键节点发现算法进行对比.实验结果表明,单形中心性可以有效刻画网络中节点的拓扑特征且KDSC算法能有效地发现网络中的关键节点.
Critical nodes identification in complex networks is significance for studying the survivability and robustness of networks.Most of the existing studies on critical nodes are based on the local environment of nodes and their impact on network functions,do not consider the global topology of networks.Persistent homology(PH)is a mathematical tool in computational topology,which can capture high-dimensional topological features and is widely used in the analysis of complex networks.In this paper,we first define the node simplex centrality based on persistent homology theory,and give a new metric to describe the importance of nodes in online social networks.Secondly,a key nodes discovery algorithm(KDSC)based on simplex centrality is proposed to obtain the importance ranking of nodes in social networks,thus systematically giving a key node metric and discovery method based on persistent homology.In the validation experiments,this paper evaluates and analyzes the node simplex centrality of real social networks,which is discussed in comparison with the traditional metrics;KDSC is used to solve the key nodes and compared with the classical key node discovery algorithm.The experimental results show that the simplex centrality can effectively characterize the topology of nodes in the network and the KDSC algorithm can effectively discover the key nodes in the network.
作者
钟慧
邱吕琳
张志坚
姜麟
李鑫阳
ZHONG Hui;QIU Lülin;ZHANG Zhijian;JIANG Lin;LI Xinyang(Faculty of Science,Kunming University of Science and Technology,Kunming 650500)
出处
《系统科学与数学》
CSCD
北大核心
2022年第8期2157-2179,共23页
Journal of Systems Science and Mathematical Sciences
基金
国家自然科学基金项目(11461037)
云南省教育厅科学研究基金项目(2017ZZX133)
云南省高校联合青年项目(2017FH001-116)资助课题.
