R help - Oct 2010 - help identifying clusters

Dear list:

I have a dataset of geographical data that looks like this example data:

dat<- data.frame( lon = c(rnorm(1000, mean=-10), rnorm(1000, mean=10),
rnorm(1000, mean=5)),
  lat = c(rnorm(1000, mean=40), rnorm(1000, mean=30), rnorm(1000, mean=0)))

plot(dat$lon, dat$lat)

My positions are clearly clustered (in this example there are 3 clusters).
Is there any R algorithm that
allows me to identify the three groups of points and allocate to each pair
the respective cluster?

Many thanks in advance

Jonas

	[[alternative HTML version deleted]]

Look at clustering task view
http://cran.r-project.org/web/views/Cluster.html

A simple way to do it is

library(cluster)
dat$cluster <- pam(dat, 3, stand=T)$clustering
plot(dat$lon, dat$lat, col=dat$cluster)


Nikhil Kaza
Asst. Professor,
City and Regional Planning
University of North Carolina

nikhil.list at gmail.com

On Oct 20, 2010, at 6:41 PM, jonas garcia wrote:
> dat<- data.frame( lon = c(rnorm(1000, mean=-10), rnorm(1000, mean=10),
> rnorm(1000, mean=5)),
>  lat = c(rnorm(1000, mean=40), rnorm(1000, mean=30), rnorm(1000,  
> mean=0)))
>
> plot(dat$lon, dat$lat)