[Thuban-list] Patch: classifiers

[Moritz Lennert]

Daniel Calvelo Aros said:
> Hi all.
>
> The attached patch provides four classifiers to thuban's available three.
>
[...]

> 2. "Distribution discontinuities": this is based on a very nice idea and
> code
> from J.P. Grimmeau of the ULB (thanks Moritz!). It will automatically find
> discontinuities in the data distribution and select breaks accordingly.
> I'm
> not sure on the proper way of giving credit, though.

If it is possible to include the mention "Jean-Pierre Grimmeau, IGEAT-ULB,
grimmeau at ulb.ac.be" in the description text, that would be great.

> 4. "Optimized distribution classes": an extra optimization pass after the
> previous clustering. The optimization is a simulated annealing applied
> with a
> target function based on Jenks & Caspall (1971). It's not really useful
> for
> now, but all the hooks are there to use it as a "natural breaks" finder.

I haven't had time to look at your patch, but one of the weak points of
the use of the Jenks algorithm in ArcView is the absence of information
about the evolution of variance with the number of classes. So, do you
think it would be possible to include a little table, showing the
proportion of intra-group variance in the total variance for a series of
solutions (i.e. 2 classes to 12 classes) as a decision-aid on the best
number of classes ?

Again, thanks a lot for your work ! These classifiers plus (hopefully one
day) proportionate circles will make it much easier for me to convince my
colleagues to switch to Thuban.

Moritz