Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.14) Gecko/20110223 Lightning/1.0b2 Thunderbird/3.1.8
Dear Listers,
I’ve a
weighted and undirected network form co-occurrence of
keywords in bibliographic records. I would like to
breakdown the network through network statistics for
visualization. I tried igraph’s clusters, subcomponent,
decompose.graph and articulation.points but nothing works.
What else I could try?
Try the community detection algorithms; "clusters", "subcomponent"
and "decompose.graph" find connected components only, and chances
are that your graph is (almost) fully connected. Community detection
algorithms look for regions that are more densely connected than
their surroundings.
Try these as a start in R:
?fastgreedy.community
?walktrap.community
?community.edge.betweenness (if your graph is not too large)
?leading.eigenvector.community
?spinglass.community