首页 | 官方网站   微博 | 高级检索  
     

基于复杂网络动力学模型的链路预测方法
作者姓名:潘永昊  于洪涛  吴翼腾
作者单位:国家数字交换系统工程技术研究中心,河南 郑州 450002
基金项目:国家自然科学基金资助项目(61803384)
摘    要:链路预测是复杂网络中研究缺失连边和未来形成连边的重要组成部分,当前基于网络结构的链路预测方法成果丰富,而基于复杂网络动力学模型的链路预测研究较少。针对无权无向网络,首先构建了复杂网络动力学模型,然后给出了基于复杂网络动力学模型的链路预测节点中心性的量化评价指标,最后通过给出的节点中心性量化指标,提出了由复杂网络动力学模型定义的链路预测方法。通过在真实网络数据集上进行的实验表明,提出的链路预测方法较基准方法有明显的预测精度的提升。

关 键 词:复杂网络  链路预测  网络动力学  

Link prediction method based on complex network dynamics model
Authors:Yonghao PAN  Hongtao YU  Yiteng WU
Affiliation:National Digital Switching System Engineering and Technological R&D Center,Zhengzhou 450002,China
Abstract:Link prediction is an important part of the study of missing links and future formations in complex networks.Currently,network structure-based link prediction methods are rich in results.Research on link prediction based on complex network dynamics model is rare.Firstly,a complex network dynamics model for unlicensed and undirected networks was constructed.Then the quantitative evaluation index of the link prediction node centrality based on the complex network dynamics model was given.Finally,the link prediction method defined by the complex network dynamics model was proposed by the given node centrality quantitative index.Experiments on real network datasets show that the proposed link prediction method has obvious prediction accuracy improvement.
Keywords:complex network  link prediction  network dynamics  
点击此处可从《》浏览原始摘要信息
点击此处可从《》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号