R help - May 2002 - Discretization of numeric attributes

Dear R-helpers:

I am interested in discretization methods for numerical attributes, as they 
are reported in the 'machine learning' community. For example, the work
of
Fayyad & Irani (IJCAI-93), Kononenko, entropy-based approaches, MDL 
principle, the C4.5 approach, etc. I am especially interested in those 
methods that take a factor as goal target into account for discretizing 
continuous-valued attributes.

Is there an implementation of any one of those methods available in the 
many libraries available for R ?

Hans Werner Borchers
ABB Corporate Research

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

On Tue, 7 May 2002, Hans W. Borchers wrote:
> Dear R-helpers:
>
> I am interested in discretization methods for numerical attributes, as they
> are reported in the 'machine learning' community. For example, the
work of
> Fayyad & Irani (IJCAI-93), Kononenko, entropy-based approaches, MDL
> principle, the C4.5 approach, etc. I am especially interested in those
> methods that take a factor as goal target into account for discretizing
> continuous-valued attributes.
>
> Is there an implementation of any one of those methods available in the
> many libraries available for R ?
rpart and tree, for example.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

>
>rpart and tree, for example.
Yes, I know that recursive partitioning does some kind of splitting 
numerical variables. I looked into 'rpart.chm', but could not identify a
function that I could call. What I would need is something like

discretize(num.var, some.factor, method="FayyadIrani, MDL, Kononenko,
C45")

giving back a factor variable with selected intervals or splitting points 
as names. Can I extract something similar from one of the packages rpart or 
tree?

Thanks in advance, Hans Werner Borchers.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._