摘要
传统簇生长法时间复杂度高、对分形标度关系刻画不够精准,且关键节点在控制网络结构和功能方面具有重要作用。为选择具有代表性的节点来分析网络自相似分形问题,提出一种基于K-shell的复杂网络簇生长法,通过K-shell分解和节点信息熵选取核心层最具影响力节点作为簇生长法的种子节点计算网络的分形维数。实验结果表明所提方法对网络的分形性质刻画得更加细致,能够计算出更加准确的分形维数。
The traditional cluster-growing method has high time complexity,inaccurate description of fractal scale relationship,and key nodes are important in controlling network structure and function.In order to select representative nodes to analyze the network self-similarity fractal problem,we propose a K-shell-based cluster-growting method of complex networks,in which the most influential nodes in the core layer are selected as the seed nodes of the cluster growth method to calculate the fractal dimension of the network through K-shell decomposition and node information entropy.Experimental results show that the proposed method can perfectly observe the fractal properties of the network and calculate the fractal dimension more accurately.
作者
张耀波
张胜
王雨萱
熊聪源
ZHANG Yaobo;ZHANG Sheng;WANG Yuxuan;XIONG Congyuan(School of Information Engineering,Nanchang HangKong University,Nanchang 330063,China)
出处
《复杂系统与复杂性科学》
北大核心
2025年第1期11-17,共7页
Complex Systems and Complexity Science
基金
国家自然科学基金(61661037)
江西省教育厅科技项目(GJJ170575)
南昌航空大学研究生创新专项基金(YC2022-053)。
