centrality_alpha: Alpha Centrality

View source: R/centrality.R

centrality_alphaR Documentation

Alpha Centrality

Description

Alpha centrality (Bonacich and Lloyd 2001) gives every node an exogenous score of one plus the scores of the nodes that point to it:

x = \alpha A^{T} x + e, \qquad x = (I - \alpha A^{T})^{-1} e.

The attenuation is fixed at \alpha = 1 and e is a vector of ones, the defaults of igraph::alpha_centrality().

Usage

centrality_alpha(x, mode = "all", ...)

Arguments

x

Network input accepted by centrality.

mode

For directed networks: "all" (default), "in" or "out".

...

Further arguments to centrality, such as directed and normalized.

Details

A holds the edge weights, or ones with weighted = FALSE, with the diagonal set to zero. The alpha argument of centrality is the weight-inversion exponent, which this measure does not read. The scores are positive when the spectral radius of A is below one and can be negative otherwise. A singular system or a negative edge weight raises an error of class cograph_singular_system.

On a directed network mode = "in" sums over incoming ties as above and equals igraph::alpha_centrality(), mode = "out" uses A in place of A^{T} and so sums over outgoing ties, and mode = "all" (default) uses the symmetrized weights A + A^{T}. On an undirected network the three modes agree.

Value

A named numeric vector with one score per node, in input node order.

References

Bonacich, P., & Lloyd, P. (2001). Eigenvector-like measures of centrality for asymmetric relations. Social Networks, 23(3), 191-201. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/S0378-8733(01)00038-7")}.

See Also

centrality_katz, centrality_eigenvector, centrality.

Examples

centrality_alpha(regulation_net)

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