XGR (version 1.1.4)

xBigraph: Function to obtain communities from a bipartitle graph

Description

xBigraph is supposed to obtain communities from a bipartitle graph.

Usage

xBigraph(g, algorithm = c("louvain", "leading_eigen", "fast_greedy",
"walktrap"), seed = 825, glayout = layout_with_kk,
edge.color.alternative = c("grey70", "grey95"), verbose = TRUE)

Arguments

g

an object of class "igraph" (or "graphNEL") for a bipartitel graph with a 'type' node attribute

algorithm

the algorithm to initialise community from a projected graph. It can be 'louvain' using multi-level optimization, 'leading_eigen' using leading eigenvector, 'fast_greedy' using greedy optimization, and 'walktrap' via short random walks

seed

an integer specifying the seed

glayout

a graph layout function. This function can be one of "layout_nicely" (previously "layout.auto"), "layout_randomly" (previously "layout.random"), "layout_in_circle" (previously "layout.circle"), "layout_on_sphere" (previously "layout.sphere"), "layout_with_fr" (previously "layout.fruchterman.reingold"), "layout_with_kk" (previously "layout.kamada.kawai"), "layout_as_tree" (previously "layout.reingold.tilford"), "layout_with_lgl" (previously "layout.lgl"), "layout_with_graphopt" (previously "layout.graphopt"), "layout_with_sugiyama" (previously "layout.kamada.kawai"), "layout_with_dh" (previously "layout.davidson.harel"), "layout_with_drl" (previously "layout.drl"), "layout_with_gem" (previously "layout.gem"), "layout_with_mds", and "layout_as_bipartite". A full explanation of these layouts can be found in http://igraph.org/r/doc/layout_nicely.html

edge.color.alternative

two alternative colors for edges within the community (grey70 by default) and edges between communities (grey95 by default)

verbose

logical to indicate whether the messages will be displayed in the screen. By default, it sets to true for display

Value

It returns an igraph object, appended by node attributes including "community" for communit memberships, "contribution" for contribution to its community, "xcoord" for x-coordinates, "ycoord" for y-coordiates, and by edge attributes including "color" for between-community edges ('grey95') and within-community edges ('grey70').

See Also

xGGnetwork

Examples

# NOT RUN {
# 1) generate a random bipartite graph
set.seed(123)
g <- sample_bipartite(100, 50, p=0.1)
V(g)$name <- V(g)

# }
# NOT RUN {
# 2) obtain its community
ig <- xBigraph(g)
gp <- xGGnetwork(ig, node.label='name', node.label.size=2,
node.label.color='black', node.label.alpha=0.8, node.label.padding=0,
node.label.arrow=0, node.label.force=0.002, node.shape='type',
node.shape.title='Type', node.xcoord='xcoord', node.ycoord='ycoord',
node.color='community', node.color.title='Community',
colormap='jet.both', ncolors=64, zlim=NULL, node.size='contribution',
node.size.range=c(1,4), node.size.title='Contribution', slim=NULL,
edge.color="color",edge.color.alpha=0.5,edge.curve=0,edge.arrow.gap=0)

## make it discrete for the colorbar
gp +
scale_colour_gradientn(colors=xColormap('jet.both')(64),breaks=seq(1,max(V(ig)$community)))
+ guides(color=guide_legend(title="Community"))
# }