[R] plotting dendrograms from cluster analyses

Philippe Grosjean phgrosje at ulb.ac.be
Tue Aug 28 14:37:46 CEST 2001


>I started a more general approch for dendrograms a year ago using
>class "dendrogram". If you coerce an hclust object to class dendrogram
>using as.dendrogram() then you can use cut.dendrogram to do what you
>want (it returns a list with the upper and lower parts of the cutted
>tree).

Yes, I have seen also some changes in the FAQ of, I think, R v. 1.2.3. I've
written code last year for v. 1.1.1 and now I have to finish the library for
the end of the year. I am a bit embarrassed with this. Should I move to
'dendrogram' objects or should I keep working on 'hclust' objects? Moreover,
my library has to be compatible with both R and Splus. Splus implementation
of hclust is somewhat different but quite close. Do you implement
'dendrogram' objects for Splus too?

I have also the same problem with time management in time series, and with
graphs ("classical" versus lattice)... R is moving so fast these time, as
does Splus, but who will complain...!


>Unfortunately this is unfinished yet, especially the plotting method
>does not print the labels, hence I didn't mention it earlier in this
>thread. I plan to finish this for R 1.4 (to be released by the end of
>the year), but don't hold your breath.

>Of course (as always) volunteers are more than welcome!

>See example(cut.dendrogram)

Thank you for the info.

By the way, I have an additional question. Does someone implement something
similar to hclust, but with "time connexity" (well, this is a rough
translation of French ;-) I mean, if you want to cluster observations of a
time serie, you could be interested by producing a dendrogram WITHOUT
reordering observations. You still keep obs 1, 2, 3,..., n in this order in
the dendrogram. But except this constraint, you apply the same algorithm.

All the best,

Philippe Grosjean


...........]<(({°<...............<°}))><...............................
 ) ) ) ) )	 __               	 __
( ( ( ( ( 	|__)              	|  _
 ) ) ) ) )	|   hilippe       	|__)rosjean
( ( ( ( ( 	Marine Biol. Lab., ULB, Belgium
 ) ) ) ) )	                  	 __
( ( ( ( ( 	|\  /|            	|__)
 ) ) ) ) )	| \/ |ariculture &	|__)iostatistics
( ( ( ( (
 ) ) ) ) )	e-mail: phgrosje at ulb.ac.be or phgrosjean at sciviews.org
( ( ( ( ( 	SciViews project coordinator (http://www.sciviews.org)
 ) ) ) ) )      tel: 00-32-2-650.29.70 (lab), 00-32-2-673.31.33 (home)
( ( ( ( (
 ) ) ) ) )      "I'm 100% confident that p is between 0 and 1"
( ( ( ( (                                  L. Gonick & W. Smith (1993)
 ) ) ) ) )
.......................................................................


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list