lans: Local Adaptive Network Sparsification

In simplifyNet: Network Sparsification

Local Adaptive Network Sparsification

Description

Remove all edges under certain probability of the fractional edge weight, alpha.

Usage

lans(network, alpha, output, directed = FALSE)

Arguments

network	Weighted adjacency matrix, weighted igraph network, or edge list formatted | n1 | n2 | weight | with colnames c("n1", "n2", "weight").
alpha	The alpha value is a predetermined threshold to designate statistically important edges by their fractional edge weight at each node. If the probability of choosing that edge via the CDF is less than or equal to alpha, then the edge is not included.
output	If the output should be directed or undirected. Default is that the output is the same as the input based on adjacency matrix symmetry. If the default is overridden, set as either "undirected" or "directed".
directed	If TRUE, specifies that the inputted network is directed. Default is FALSE.

Details

For more information on finding alpha values, see: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0016431#s5

Value

Weighted adjacency matrix of sparsified network.

Author(s)

Andrew Kramer

Alexander Mercier

References

Foti, N. J., Hughes, J. M., & Rockmore, D. N. (2011). Nonparametric sparsification of complex multiscale networks. PloS one, 6(2), e16431.

Examples

#Generate random ER graph with uniformly random edge weights
g = igraph::erdos.renyi.game(100, 0.1)
igraph::E(g)$weight <- runif(length(igraph::E(g)))
#Sparsify g via LANS
S = lans(g, alpha = 0.3, output = "undirected", directed = FALSE)
#Convert sparsifier to edge list
S_List = simplifyNet::Mtrx_EList(S, directed = FALSE)
sg = simplifyNet::net.as(S_List, net.to="igraph", directed=FALSE)
igraph::ecount(sg)/igraph::ecount(g)#fraction of edges in the sparsifier

simplifyNet documentation built on Nov. 15, 2022, 5:05 p.m.