Fusing data depth with complex networks: Community detection with prior information

Yahui Tian and Yulia R. Gel

Computational Statistics & Data Analysis, 2019, vol. 139, issue C, 99-116

Abstract: A new nonparametric supervised algorithm is proposed for detecting multiple communities in complex networks using the Depth vs. Depth (DD(G)) classifier. The key idea behind the new clustering method is the notion of robust and data-driven data depth methodology that still remains new and unexplored in network sciences. The developed new DD(G)-method is inherently geometric and allows to simultaneously account for network communities and outliers. Although the data-based classifier operates within a supervised learning framework, the related nonparametric notion of depth in networks can be used in a more general context, including (semi) supervised and unsupervised learning. Utility of the new approach is illustrated by using the benchmark political blogs data, “dark” terrorist networks, and analysis of bill cosponsorship in the Italian Parliament.

Keywords: Community detection; Sparse complex networks; Data depth; Outliers (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167947319300088
Full text for ScienceDirect subscribers only.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:csdana:v:139:y:2019:i:c:p:99-116

DOI: 10.1016/j.csda.2019.01.007

Access Statistics for this article

Computational Statistics & Data Analysis is currently edited by S.P. Azen

More articles in Computational Statistics & Data Analysis from Elsevier
Bibliographic data for series maintained by Catherine Liu ().