Informative core identification in complex networks

Informative core identification in complex networks
复制标题

复杂网络中的信息核心识别

DOI:
10.1093/jrsssb/qkac009
复制
发表时间:
2023
期刊:
Journal of the Royal Statistical Society Series B: Statistical Methodology
影响因子:
--
通讯作者:
Li, Tianxi
Li, Tianxi
中科院分区:
--
文献类型:
--
作者:
Miao, Ruizhong;Li, Tianxi

文献摘要

参考文献

相似文献

在复杂网络中,具有有趣结构的核心组件通常隐藏在非信息连接中。非信息分量引入的噪声和偏差会掩盖显著结构,限制许多网络建模过程的有效性。本文介绍了一种新的核心-外围模型的非信息网络的外围结构,而不强加一个特定形式的核心。我们提出了谱算法的核心识别一般下游网络分析任务下的模型。该算法享有强大的性能保证,并可扩展的大型网络。我们通过广泛的模拟研究来评估这些方法,这些研究证明了它们比多个传统的核心-外围方法具有优势。该方法还用于从引文网络中提取核心结构,这导致更可解释的分层社区检测。
In a complex network, the core component with interesting structures is usually hidden within noninformative connections. The noises and bias introduced by the noninformative component can obscure the salient structure and limit many network modeling procedures’ effectiveness. This paper introduces a novel core–periphery model for the noninformative periphery structure of networks without imposing a specific form of the core. We propose spectral algorithms for core identification for general downstream network analysis tasks under the model. The algorithms enjoy strong performance guarantees and are scalable for large networks. We evaluate the methods by extensive simulation studies demonstrating advantages over multiple traditional core–periphery methods. The methods are also used to extract the core structure from a citation network, which results in a more interpretable hierarchical community detection.
DOI: --
发表时间: 2017-09
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
A. Athreya;D. E. Fishkind;M. Tang;C. Priebe;Youngser Park;J. Vogelstein;Keith D. Levin;V. Lyzinski;Yichen Qin;D. Sussman
通讯作者: A. Athreya;D. E. Fishkind;M. Tang;C. Priebe;Youngser Park;J. Vogelstein;Keith D. Levin;V. Lyzinski;Yichen Qin;D. Sussman
DOI: 10.1051/ps/2014017
发表时间: 2013
期刊: arXiv: Statistics Theory
影响因子: --
作者:
C. Butucea;Yu. I. Ingster;I. Suslina
通讯作者: I. Suslina
DOI: 10.1093/biomet/asaa006
发表时间: 2020-06-01
期刊: BIOMETRIKA
影响因子: 2.7
作者:
Li, Tianxi;Levina, Elizaveta;Zhu, Ji
通讯作者: Zhu, Ji