Re: [igraph] Comparing community structures

[Tadaishi Yatabe-Rodriguez]

I've been trying to get this code running to get the agreement between different runs of the Infomap algorithm

info <- list()

for (i in 1:10){

info[[i]]<-infomap.community(el.13, nb.trials=100)

}

### Which works fine, but for comparing the community structures I use this for loop:

nmi <- rep(NA, 45)

for (i in 1:10) {for (n in (i+1):10) { for (j in 1:45) {

nmi[j] <- compare(info[[i]], info[[n]], method="nmi")

}}}

which just gives me 1's.

Can someone get this code running?

Thanks,

Tada

On Thu, Sep 11, 2014 at 1:09 PM, Tadaishi Yatabe-Rodriguez <address@hidden> wrote:

Hi,

I'm running the Infomap algorithm, but since its random by nature, I'd like to get an idea of how alike community structures are between realizations of the algorithm. I''ve used community structure comparison algorithms such as nmi in the compare function, but I've noticed that I would need to do some sort of pairwise comparison of the community objects I've created, which I'm not quite sure how to code.

Also, Tamas has pointed put to me that I could create a matrix that counts the number of times a pair of nodes is in the same cluster and use that as an indicator of the most likely clusters. Then again, I'm not quite sure how to do this.

Thanks,

Tada

--

Tadaishi Yatabe

DVM, MPVM, PhD (C)

Center for Animal Disease Modeling and Surveillance (CADMS)

Department of Medicine and Epidemiology

School of Veterinary Medicine

University of California Davis

--

Tadaishi Yatabe

DVM, MPVM, PhD (C)

Center for Animal Disease Modeling and Surveillance (CADMS)

Department of Medicine and Epidemiology

School of Veterinary Medicine

University of California Davis