b_graph.affiliation: Simulation of an Affiliation Graph

In mixer: Random Graph Clustering

Description

Simulate an affiliation network with a given number of clusters, specific class proportions and within/between connection probabilities.

Usage

graph.affiliation( n=100, alphaVect=c(1/2,1/2), lambda=0.7,
                   epsilon=0.05, directed=FALSE ) 

Arguments

n	number of nodes of the simulated network.
alphaVect	vector of cluster proportions.
lambda	within-cluster probability of edge.
epsilon	between-clusters probability of edge.
directed	TRUE for directed graphs.

Details

graph.affiliation simulates a simple Erd<c3><b6>s-R<c3><a9>nyi Mixture of Graph model, using the same within-cluster edge probability for all clusters and a unique between-cluster edge probability.

Value

graph.affiliation returns a list of 2 objects:

x	an adjacency matrix of size n by n,
cluster	a vector of integers indicating the cluster to which each node is allocated.

Author(s)

Christophe Ambroise

Examples

graph.affiliation(n=100,c(1/3,1/3,1/3),0.8,0.2)->g
str(g)

mixer documentation built on Feb. 21, 2018, 1:02 a.m.