Characterizing variability in complex network community structure with a recursive significance clustering scheme

Network science has presented community detection as a valuable tool for revealing the functional modules in complex systems as rooted in the wiring architectures of complex networks. The varying procedures of community detection can produce, however, divisions of a network into communities that are...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Economou, Karsten N, Norman, Cassie R, Gentleman, Wendy C
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Network science has presented community detection as a valuable tool for revealing the functional modules in complex systems as rooted in the wiring architectures of complex networks. The varying procedures of community detection can produce, however, divisions of a network into communities that are prone to degeneracy given small changes to the network's configuration. This yields network partitions of similar merit that are structurally dissimilar, which is especially problematic when the network is constructed on uncertain data. To reconcile with the ambiguity in interpreting degenerate network partitions as representations of underlying system function, we introduce a recursive significance clustering scheme that identifies the subsets of nodes that have stable joint community assignments under network perturbation. These robust node groups are referred to here as cores, and represent well-supported features of the network. We show that cores characterize the variability inherent to non-overlapping community structure in weighted and directed networks and indicate robust community assignments that are cohesive under temporal evolution of the network.
DOI:10.48550/arxiv.2409.12852