I haven't known of KEEL until I've read your mail. I'm missing a proper
documentation of the framework to use it, even if the frameworks
suits better for my problem.
Currently I'm working on a representation of a rule as a strong typed
syntax tree, which also works quite well until now. Don't know how it
will work on complex data mining rules.
Are there some more detailed documentation and tutorials for KEEL? While
my project is still in an early state, changing the framework is still
an easy task and if I've more informations about the functions I could
start directly in testing the framework for my problem.
----- Ursprüngliche Mail -----
Von: Uday kamath <[log in to unmask]>
An: [log in to unmask]
Gesendet: Sat, 20 Apr 2013 20:42:35 +0200 (CEST)
Betreff: Re: Evolve Rules with ECJ
Have you looked at KEEL ? http://www.keel.es/
It has more machine learning and Rule Implementations and some specifics
for Pitt and Michigan approach. Hope that helps
On Thu, Apr 18, 2013 at 5:11 PM, Bojan Janisch <
[log in to unmask]> wrote:
> Hm... That does not sound good. Wouldn't it be possible to use the
> GP-package instead and define different types of nodes as different parts
> of rulecode?
> I don't want to start from scratch, I don't know how long it will take to
> create a new representation for my problem. I thought the rule package is
> so abstract created that it could contain anything?
> Von: Sean Luke
> Gesendet: 18.04.2013 21:45
> An: [log in to unmask]
> Betreff: Re: Evolve Rules with ECJ