交叉抽样在复杂网络中的研究与应用  

Research and Application of Cross Sampling on Complex Network

作　　者：刘胜久 伍小兵[1] 曹小平[2] 汪应[1] 欧明辉 Liu Shengjiu;Wu Xiaobing;Cao Xiaoping;Wang Ying;Ou Minghui(Big Data and Internet of Things School,Chongqing Vocational Institute of Engineering,Chongqing 402260,China;School of Artifical Intelligence,Chongqing Creation Vocational College,Chongqing 402160,China)

机构地区：[1]重庆工程职业技术学院大数据与物联网学院,重庆402260 [2]重庆科创职业学院人工智能学院,重庆402160

出　　处：《南京师范大学学报（工程技术版）》2023年第1期84-92,共9页Journal of Nanjing Normal University(Engineering and Technology Edition)

基　　金：重庆市高校创新研究群体项目(CXQT21032);重庆市自然科学基金项目(cstc2021jcyj-msxmX0532);重庆市教育委员会科学技术研究计划项目(KJQN202103404、KJQN202005403、KJQN202003409、KJQN202103401、KJZDM202203401);重庆市高等教育教学改革研究重点项目(202182);重庆市高等职业教育教学改革研究项目(Z212026)。

摘　　要：针对传统网络抽样主要是对复杂网络的节点及边进行独立抽样,提出对复杂网络的节点或边进行独立的2次抽样,再对得到的抽样网络进行分析,从而推算出原始网络的各项参数.在交叉抽样中,分析了点交叉抽样、边交叉抽样及混合抽样中的点混合抽样与边混合抽样4种交叉抽样方法,并在经典的ER、WS及BA网络模型上进行了验证.结果表明,通过交叉抽样可较好地推算出原始网络的平均度、平均路径长度、网络直径、传递聚集系数、WS聚集系数、网络维数等参数,且点混合抽样的效果最优.Among the analysis and research of complex network,in view of the traditional network sampling being mainly on sample the nodes and edges of complex network independently,this paper firstly proposes cross sampling by sample the nodes or edges of the complex network twice independently,and then calculate the parameters of the original network by sampling network.On cross sampling,four cross sampling methods including point cross sampling,edge cross sampling,point mixed sampling and edge mixed sampling are analyzed and verified on ER,WS and BA network models.The results show that the average degree,average path length,network diameter,transitivity clustering coefficient,WS clustering coefficient,and network dimension of the original network can be calculated by cross sampling,while that the point mixed sampling is the best cross sampling method.

关 键 词：复杂网络 网络抽样 交叉抽样 混合抽样 网络参数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]