[R] plotting dendrograms from cluster analyses

[Friedrich Leisch]

>>>>> On Tue, 28 Aug 2001 14:37:46 +0200,
>>>>> Philippe Grosjean (PG) wrote:

  >> 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?

If you need code or results in the near future I'd stick with hclust
if I were you. dendrogram is currently more of an experiment for more
general trees than hclust allows (more than one child nodes etc.)
... I won't replace the existing hclust, rpart classes etc. but will
provide additional plotting, printing and cutting facilities.

  > 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?

Hmm, I see no reasons why they should not work with Splus, currently
everything is written in S. If I move stuff to C for speedup things
might get different. But even then I'd probably try to be compatible.


  > 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...!

:-)

.f
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._