signnet (version 0.6.0)

signed_blockmodel: Blockmodelling for signed networks

Description

Finds blocks of nodes with intra-positive and inter-negative edges

Usage

signed_blockmodel(g, k, alpha = 0.5, annealing = FALSE)

Arguments

g

igraph object. Must have a "sign" edge attribute.

k

number of blocks

alpha

see details

annealing

logical. if TRUE, use simulated annealing (Default: FALSE)

Value

numeric vector of block assignments and the associated criterion value

Details

The function minimizes P(C)=$\alpha$N+(1-$\alpha$)P, where N is the total number of negative ties within plus-sets and P be the total number of positive ties between plus-sets. This function implementes the structural balance model. That is, all diagonal blocks are positive and off-diagonal blocks negative. For the generalized version see signed_blockmodel_general.

References

Doreian, Patrick and Andrej Mrvar (2009). Partitioning signed social networks. Social Networks 31(1) 1-11

Examples

# NOT RUN {
library(igraph)

g <- sample_islands_signed(10,10,1,20)
clu <- signed_blockmodel(g,k = 10,alpha = 0.5)
table(clu$membership)
clu$criterion

# Using simulated annealing (less change of getting trapped in local optima)
data("tribes")
clu <- signed_blockmodel(tribes,k = 3,alpha=0.5,annealing = TRUE)
table(clu$membership)
clu$criterion
# }