MathGroup Archive 2008

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

Search the Archive

Extracting information from Agglomerate.

  • To: mathgroup at
  • Subject: [mg87470] Extracting information from Agglomerate.
  • From: "W_Craig Carter" <ccarter at>
  • Date: Fri, 11 Apr 2008 01:45:02 -0400 (EDT)

I am beginning to explore Agglomerate.  I'd like to  extract information
about which elements of
a cluster are  indicated by the  dissimilarity measure. (I am fairly new to
discrete math)

For example,

alist = RandomReal[{0, 100}, {5, 3}];
Agglomerate[alist, Linkage -> "Single"]

For any cluster with non-trivial subclusters,
is it possible to extract the particular points of the subclusters that are
producing the nearest two points?  It seems it must be computed, but hidden.
I'd like to get those points and not have to traverse the subclusters to
find them.

Thanks, Craig

W. Craig Carter

  • Prev by Date: Re: pdf and accents
  • Next by Date: Re: max and min
  • Previous by thread: Re: PolarPlot
  • Next by thread: what does Method->"HighDimensionalEmbedding" actually do for GraphPlot?