多层网络节点重要性排序研究  

Research on node importance ranking of multiplayer network

作　　者：王娟 石磊[1] 王祖俪 Wang Juan;Shi Lei;Wang Zuli(School of Cybersecurity,Chengdu University of Information Technology,Chengdu 610225,China;Network ＆ Data Security Key Laboratory of Sichuan Province,University of Electronic Science ＆ Technology of China,Chengdu 610054,China)

机构地区：[1]成都信息工程大学网络空间安全学院,成都610225 [2]电子科技大学网络与数据安全四川省重点实验室,成都610054

出　　处：《计算机应用研究》2018年第6期1742-1746,共5页Application Research of Computers

基　　金：四川省科技厅应用基础研究计划资助项目(2014JY0071);网络与数据安全四川省重点实验室开放课题(NDS2015-01)

摘　　要：如何衡量节点的重要性是网络科学的基础问题之一。现有研究多集中于单层网络衡量指标的设计与改进,只能给出特定关系网络、特定测度下的结果。而节点往往存在于多种不同关系之中。如何综合衡量多层网络节点的重要性是一个更为现实而有意义的问题。提出了一个多层网络节点重要性排序框架,其核心是每层提取多个指标,对指标和层次赋权叠加构成一个综合评价矩阵,在这个矩阵基础上进行排序。实验证明,该框架可以灵活适应不同的应用场景、不同指标集合。框架给出的综合排序结果有效避免了单层网络排序的片面性。How to measure the node importance is one of the foundation issue of network science. Existing researches focus on the design and improvement of monolayer network indexes which give partial results for relationship and specific measure.However,a node often exists in various relations and how to rank a multiplayer network node is more realistic and meaningful.This paper proposed a multiplayer network node ranking framework. It extracted multi-indexes of monolayer and weighted the indexes and monolayers to get an overlay comprehensive evaluation matrix. Finally,it ranked nodes based on this evaluation matrix. The simulations show that the framework is flexible to adapt to different application scenarios,and different indexes collection. Furthermore, the framework give more comprehensive results and avoid the partiality of existing methods effectively.

关 键 词：节点重要性排序 多层网络 层次分析法 逼近理想排序法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]