April 2013


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Bojan Janisch <[log in to unmask]>
Reply To:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Mon, 22 Apr 2013 23:36:55 +0200
text/plain (51 lines)
Hi Uday,

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.


----- Urspr√ľngliche 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 ?
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