[Rd] fast nearest-neighbor in R?
nikko at hailmail.net
Thu Nov 6 02:09:05 MET 2003
It is in the works. Take a look at the archives in
https://www.stat.math.ethz.ch/mailman/listinfo/r-sig-geo for the
r-sig-geo mailing list.
For the moment you can use the k-nearest-neighbor code in Spdep which
is adapted from Ripley's knn classifier code. It uses brute force and is
as fast as algorithms based on kd or quad trees for larger data sets. For
data sets you probably won't notice the difference.
On Wed, 5 Nov 2003 08:58:17 -0800 (PST), "Mirka Zednikova"
<zednikova at yahoo.com> said:
> Is fast nearest-neighbor functionality available in R?
> I was thinking of something along the lines of what's
> currently in S+SPATIALSTATS.
> Thanks for any information anyone might have on this.
> - MZ
> R-devel at stat.math.ethz.ch mailing list
More information about the R-devel