anyone using AnswerTree 2.1?

[Tjen-Sien Lim]

If you do, I'd like to hear what you have to say about the product. I'm
completing my review of Windows classification tree software packages. Some
results are available at

http://www.recursive-partitioning.com/review.html

Thanks.

--
Tjen-Sien Lim
ts...@recursive-partitioning.com
www.Recursive-Partitioning.com
____________________________________________________________________
Get your free Web-based email! http://recursive-partitioning.zzn.com

[Frank THOMAS]

Tjen-Sien,
I am no mathematician but I always feel uneasy if I don't understand the
procedure they are proposing. And this is the case with AT. The handbook is
a sort of extended click-and-see online manual, not a real explanation of
what you should do if you habe such or such a problem.

Then, from a user view point, the trees are VERY large, difficult to see at
the screen and difficult to print. There is no way to post-process the image
within AT or to export it in third-party drawing programs.

The idea is ok, the implementation lacks finish.

Tjen-Sien Lim a écrit dans le message
<7tudet$r5c$1...@bgtnsc03.worldnet.att.net>...

>If you do, I'd like to hear what you have to say about the product. I'm
>completing my review of Windows classification tree software packages. Some

Frank Thomas
CNET/DIH/UCE
France

Le message n'engage que de la résponsabilité de son auteur et en aucun cas
celui de son employeur.

[Tjen-Sien Lim]

In article <7tvelt$qf...@news.issy.cnet.fr>,
frank....@cnet.francetelecom.fr says...

>
>Tjen-Sien,
>I am no mathematician but I always feel uneasy if I don't understand the
>procedure they are proposing. And this is the case with AT. The handbook is
>a sort of extended click-and-see online manual, not a real explanation of
>what you should do if you habe such or such a problem.

It's not unique to AT only. To understand the algorithms, you need to read
the papers mentioned in the bibliography. The current trend which I really
hate is not to publish the algorithm in any scientific journal. If you can
evaluate the various data mining software in the market, you'll find that you
can never know how the methods work. Everything becomes proprietary.

>Then, from a user view point, the trees are VERY large, difficult to see at
>the screen and difficult to print. There is no way to post-process the image
>within AT or to export it in third-party drawing programs.

I have the same problem. You can export the diagram manually to allCLEAR.
It's just very tedious to do so.

>The idea is ok, the implementation lacks finish.

I concur.

>Tjen-Sien Lim a écrit dans le message
><7tudet$r5c$1...@bgtnsc03.worldnet.att.net>...
>>If you do, I'd like to hear what you have to say about the product. I'm
>>completing my review of Windows classification tree software packages. Some
>
>
>Frank Thomas
>CNET/DIH/UCE
>France
>
>Le message n'engage que de la résponsabilité de son auteur et en aucun cas
>celui de son employeur.

--