Type: Package
Title: S-Core Graph Decomposition
Version: 0.1.2
Maintainer: Christos Adam <econp266@econ.soc.uoc.gr>
Description: S-Core Graph Decomposition algorithm for graphs. This is a method for decomposition of a weighted graph, as proposed by Eidsaa and Almaas (2013) <doi:10.1103/PhysRevE.88.062819>. The high speed and the low memory usage make it suitable for large graphs.
License: GPL-3
LinkingTo: Rcpp
Encoding: UTF-8
URL: https://github.com/cadam00/scoredec, https://cadam00.github.io/scoredec/
BugReports: https://github.com/cadam00/scoredec/issues
Imports: Rcpp (≥ 1.0.12), Rfast, igraph
Suggests: Rfast2, knitr, rmarkdown, testthat (≥ 3.0.0)
Config/testthat/edition: 3
RoxygenNote: 7.3.2
VignetteBuilder: knitr, rmarkdown
NeedsCompilation: yes
Packaged: 2024-11-20 06:21:59 UTC; Administrator
Author: Christos Adam [aut, cre]
Repository: CRAN
Date/Publication: 2024-11-20 06:50:01 UTC

s-core community decomposition

Description

s-core community decomposition

Usage

s_coreness(g = NULL, W = NULL, mode = "all")

Arguments

g

igraph object. It is a weighted graph. If it is not weighted, then the igraph::coreness function will be used. It can be used as an alternative to W.

W

matrix object. It is an adjacency matrix. It can be used as an alternative to g.

mode

character object. It can be one of "all", "in" or "out".

Details

s-core community decomposition implementation. Only one of g or W must be provided.

While the source code is not as clear as the one at brainGraph::s_core, it is very speed and memory efficient. In case that the adjacency matrix W is provided instead of the graph g is provided, then this function is very speed and memory efficient.

Note that in cases that the adjacency matrix W is known to be symmetric (checked, for example, with base::isSymmetric or Rfast::is.symmetric), then mode = "in" and mode = "out" will produce the same result more efficiently. For efficiency reasons not checking it is chosen, but user should do it.

Value

Integer vector with s-coreness attribute to each vertex.

References

Eidsaa, M. and Almaas, E. (2013) ‘s-core network decomposition: A generalization of k-core analysis to weighted networks’, Phys. Rev. E., American Physical Society, 88, 062819. doi:10.1103/PhysRevE.88.062819.

Watson, C.G. (2024). brainGraph: Graph Theory Analysis of Brain MRI Data. R package version 3.1.0. doi:10.32614/CRAN.package.brainGraph.

Examples

set.seed(42)

## Create a dummy symmetric adjacency matrix
n <- 5
W <- matrix(runif(n^2),n)
W[lower.tri(W)] = t(W)[lower.tri(W)]
diag(W) <- 0

print(scoredec::s_coreness(g = NULL, W = W, mode = "all"))
#> [1] 3 1 2 4 4

base::isSymmetric(W)
#> [1] TRUE

all.equal(scoredec::s_coreness(g = NULL, W = W, mode = "all"),
scoredec::s_coreness(g = NULL, W = W, mode = "in"))
#> [1] TRUE

# Create a dummy undirected graph
g <- igraph::graph_from_adjacency_matrix(adjmatrix = W,
                                         mode      = "undirected",
                                         weighted  = TRUE)
print(scoredec::s_coreness(g = g, W = NULL, mode = "all"))
#> [1] 3 1 2 4 4