Not for hclust() since it provides results for all clusters from 1 to n (the
number of observations). Adding a point can change the definition of the
clusters. You could use cutree() to assign the observations to clusters for a
particular number of clusters, but then you must decide what rule to use in
assigning your new point to one of those clusters (the method= argument in
hclust). A simple solution would be to identify to which of the original points,
your new point is closest. Assign the new point to the cluster that point is in.
Another would be to use aggregate() to compute the centers of the clusters and
assign the new point to the closest center. These two approaches will not
necessarily agree with one another.
-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352
-----Original Message-----
From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of TJUN KIAT TEO
Sent: Tuesday, October 11, 2016 2:57 AM
To: r-help at r-project.org
Subject: [R] Hclust
For the hclust function in R, is there a predict function that would work to
tell me which cluster does a new observation belong to? Same question for dbscan
and self organizing map
Thanks
Tjun Kiat Teo
[[alternative HTML version deleted]]
______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.