Services & Resources / Wolfram Forums / MathGroup Archive
-----

MathGroup Archive 2008

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: very long dendrogram

  • To: mathgroup at smc.vnet.net
  • Subject: [mg91444] Re: [mg91422] very long dendrogram
  • From: Curtis Osterhoudt <cfo at lanl.gov>
  • Date: Fri, 22 Aug 2008 03:11:08 -0400 (EDT)
  • Organization: LANL
  • References: <200808210815.EAA07234@smc.vnet.net>
  • Reply-to: cfo at lanl.gov

Hi, Paul, 

    I have been doing something similar, though not with quite that number of 
leaves for the dendrograms. I think you'll want to do something like the 
following (setting AspectRatio and possibly ImageSize):

DendrogramPlot[Rest /@ sheet1dat,  AspectRatio -> 8, 
 ImageSize -> {Automatic, 2500}, LeafLabels -> sheet1sites]

  I've never had a problem with the LeafLabels being out of order or anything 
like that. Note that my "sheet1sites" IS simply a list (I possibly made the 
right choice of not paying too much attention to the documentation, or it's 
wrong, or it didn't apply to my case or something), and the result is quite 
pretty.

   Note that it's also possible (through some really kludgy hackery) to 
present these things in what I call "radial" form, quite like the plot about 
halfway down this page: http://www.santafe.edu/~aaronc/randomgraphs/ . 
Basically, I make the regular, rectangular dendrogram first, then do a 
conformal transform on all the line coordinates to get the radial "spokes" 
and circumferential "branches", then plot the result. It takes a while (and 
probably longer on your dataset, as there's lots of interpolation my code 
does to get smoothly-curved circular arcs), but it's quite nice. The leaves 
may also be labeled and color-coded, in a way that's nice and readable.
   I'm not quite sure I have perfected the package yet (and it may be 
considered somewhat proprietary by a colleague at this point), but I'll be 
happy to send examples off-list.

                Best wishes, 
                           Curtis O.


On Thursday 21 August 2008 02:15:09 Paul Slater wrote:
> I have produced a hierarchical clustering
> of 3,107 U. S. counties based on 1995-2000
> migration flows between the counties using a certain non-standard
> methodology
> (http://arxiv.org/abs/0807.1550). I have been able to
> convert the results into a cluster object for use by
> the `Hierarchical Clustering" package of Mathematica.
>
> However, if I try to produce the full (very long)
> dendrogram or even use the command, say,
> TruncateDendrogram->{1,25}], , using say
> Orientation->Right, the result is
> not readable, being "crammed" into a small space.
> So, I would like to expand this space by several orders.
>
> If this problem can be overcome, I would also
> like to be able to eventually  label the leaves of the dendrogram
> by the names of the counties. But
> "For  Cluster objects, the  LeafLabels option
> can take a function but not a list or  Automatic,
> as there is no unambiguous mapping between
> the labels and data points in a  Cluster object."
> So, I am not sure how to proceed in this regard.
>
> Paul B. Slater



-- 
==========================================================
Curtis Osterhoudt
cfo at remove_this.lanl.and_this.gov
PGP Key ID: 0x4DCA2A10
Please avoid sending me Word or PowerPoint attachments
See http://www.gnu.org/philosophy/no-word-attachments.html
==========================================================


  • Prev by Date: Re: Re: Solving nonlinear inequality constraints
  • Next by Date: Re: Minimize with complex values
  • Previous by thread: Re: very long dendrogram
  • Next by thread: Re: very long dendrogram