Re: [igraph] clustering using igraph

[Javier Rodriguez Zaurin]

Hi Hong Cai

Well, speaking in general terms, igraph has a cluster method that you might find useful: http://www.inside-r.org/packages/cran/igraph/docs/no.clusters

Also you might want to have a look to: http://stats.stackexchange.com/questions/139490/approach-and-example-of-graph-clustering-in-r 

which uses the fastgreedy algorithm to find communities. 

If you are going to be using igraph+R heavily I would recommend having a look to this book. I found it fantastic: http://www.amazon.co.uk/Statistical-Analysis-Network-Data-Use/dp/1493909827/ref=sr_1_1?ie=UTF8&qid=1458316128&sr=8-1&keywords=statistical+analysis+of+network+data+with+r

Good luck!

J.

2016-03-18 15:42 GMT+00:00 Hong Cai <address@hidden>:

Dear all,

 

I have a question regarding how to use igraph to do the clustering in R. I am new R and igraph user. I very appreciate your help.

 

For example, I have 4 nodes {A,B,C,D}, some of the nodes are connected while some are not, for example,{(A,B),(B,C),(C,D)}

For each node, I assign a value to it. I want to use K mean method to get two clusters. For each cluster, nodes should be connected with at least one node in the same cluster. Therefore, in this example, Node A and D cannot be grouped as a cluster as they are not connected.

 

 

Thank you very much for your attention and time.

Best regards,

Hong

_______________________________________________
igraph-help mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/igraph-help