[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wekalist
Subject:    Re: [Wekalist] Filters and bagging
From:       Peter Reutemann <fracpete () waikato ! ac ! nz>
Date:       2006-04-24 22:30:29
Message-ID: 444D5185.3050307 () waikato ! ac ! nz
[Download RAW message or body]

> I have a training set and i have the classes avaliable in nominal form. 
> i was wondering if i should use unsupervised discretize filter or 
> supervised discretize filter. Can the unsupervised discretize filter be 
> used in supervised learning?

Supervised means only that the filter will take the class attribute and 
therefore its distribution into account, the unsupervised will just 
ignore it (that's why you have to specify the number of bins for the 
Discretize filter yourself). In other words, yes, you can use the 
unsupervised one.

> I would also like to know if anyone could provide me more information on 
> the IBk algorithm.

IBk paper:
D. Aha, D. Kibler (1991). Instance-based learning algorithms. Machine 
Learning. 6:37-66.

HTH

Cheers, Peter
-- 
Peter Reutemann, Dept. of Computer Science, University of Waikato, NZ
http://www.cs.waikato.ac.nz/~fracpete/     +64 (7) 838-4466 Ext. 5174

_______________________________________________
Wekalist mailing list
Wekalist@list.scms.waikato.ac.nz
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
[prev in list] [next in list] [prev in thread] [next in thread]