universal: Compute universal threshold backbone

In KGodard1/Fastball-SourceCpp: Extracts the Backbone from Weighted Graphs

Description

'universal' returns a backbone graph in which edge weights are set to 1 if above the given upper parameter threshold, set to -1 if below the given lower parameter threshold, and are 0 otherwise.

Usage

universal(M, upper = NULL, lower = NULL, bipartite = NULL, narrative = FALSE)

Arguments

M	graph: Graph object of class matrix, sparse matrix, igraph, edgelist, or network object. Any rows and columns of the associated bipartite matrix that contain only zeros are automatically removed before computations.
upper	Real, FUN, or NULL: upper threshold value or function to be applied to the edge weights. Default is NULL.
lower	Real, FUN, or NULL: lower threshold value or function to be applied to the edge weights. Default is NULL.
bipartite	Boolean: TRUE if bipartite graph, FALSE if weighted graph. Default is NULL. If TRUE, input graph should be unweighted.
narrative	Boolean: TRUE if suggested text for a manuscript is to be returned

Details

If both 'upper' and 'lower' are 'NULL', a weighted projection is returned.

If 'bipartite' is 'NULL', the function tries to guess at whether the data is bipartite or unipartite based on its shape.

Value

backbone, a list(backbone, summary). The 'backbone' object is a graph object of the same class as M. The 'summary' contains a data frame summary of the inputted matrix and the model used including: model name, number of rows, number of columns, and running time.

Examples

test <- universal(davis%*%t(davis), upper = function(x)mean(x)+sd(x), lower=function(x)mean(x))
test2 <- universal(davis, upper = function(x)mean(x)+2*sd(x), lower = 2, bipartite = TRUE)
test3 <- universal(davis, upper = 4, lower = 2, bipartite = TRUE)

KGodard1/Fastball-SourceCpp documentation built on Dec. 18, 2021, 2:35 a.m.