community.licodIteratif: Iteratif Licod community detection algorithm
In Issamfalih/MUNA: MUNA: A Multiplex Network Analysis Library

Description Usage Arguments Value Author(s) References Examples

This applies the iterative version of the Licod algorithm as described in VCSJ 2014 (Yakoubi and Kanawati, 2014)

1	community.licodIteratif(graph, sigma, centrality1, centrality2, delta, vote_fun, sim_fun, memb_fun, verbose, max_iter)

`graph`	: The input igraph graph
`sigma`	: is a threshold in [0; 1]. It is used to know if a node is a leader.
`centrality1`	: is a topological centrality in graphs as degree, closeness, betweenness...
`centrality2`	: is a topological centrality in graphs as degree, closeness, betweenness...
`delta`	: is a threshold in [0; 1]. Two leaders are linked if their topological similarity is above delta.
`vote_fun`	: The vote function that should be used to compute the membership vector.
`sim_fun`	: is the similarity function. i.e similarity.invlogweighted, similarity.jaccard, similarity.dice,...
`memb_fun`	: is the membership function i.e mean, sum, ...
`verbose`	: verbose
`max_iter`	: max number of iterations if not stablized

returns an igraph communities object, please see igraph manual page for details.

Issam Falih <issam.falih@lipn.univ-paris13.fr>

Yakoubi, Z. et R. Kanawati (2014). Licod : Leader-driven approaches for community detection. Vietnam Journal of Computer Science - Springer 1, 241-256.

g <- graph.famous("Zachary")
wt <- community.licodIteratif(g)
 V(g)$color <- wt$membership+1
 g$layout <- layout.fruchterman.reingold
 plot(g, vertex.label=NA)