centrality_s_shell: s-shell Index

View source: R/centrality-batch8.R

centrality_s_shellR Documentation

s-shell Index

Description

The s-shell index (Liu, Tang, Do and Hui 2017) peels the network like the k-shell decomposition but by a strength computed from the topology. Each link gets the asymmetric weight

w_{ij} = 1 + (k_i\, k^{out}_j)^a,

where k^{out}_j counts the neighbors of j outside the closed neighborhood of i, and the strength of i is s_i = \sum_{j \in N(i)} w_{ij}. At each step the nodes at the minimum remaining strength are removed, the removals cascade, and the removed nodes receive the next shell index.

Usage

centrality_s_shell(x, s_shell_a = 0.5, ...)

Arguments

x

Network input accepted by centrality.

s_shell_a

Exponent a of the link weights (default 0.5, the value the source recommends).

...

Further arguments to centrality.

Details

Direction, edge weights and self-loops are ignored. The index is an ordinal counter starting at 1 for the outermost shell, so values are comparable within a network only. Higher indices mark more central nodes. Isolated nodes form shell 1 on their own and shift every other shell up by one. With a = 0 the shells are the dense ranks of the k-core numbers. A s_shell_a that is not a single non-negative number raises an error of class cograph_bad_parameter.

Value

A named integer vector with one shell index per node, in input node order.

References

Liu, Y., Tang, M., Do, Y., & Hui, P. M. (2017). Accurate ranking of influential spreaders in networks based on dynamically asymmetric link weights. Physical Review E, 96(2), 022323.

See Also

centrality_coreness, centrality_weighted_kshell, centrality.

Examples

centrality_s_shell(regulation_net)

cograph documentation built on Nov. 19, 2026, 5:07 p.m.