I suggest you following papers for Apriori:
"Fast discovery of association rules" Agrawal et al.
"Fast algorithms for mining association rules il large db" Agrawal et al.
"Mining generalized association rules" Agrawal et al.
About Tertius I suggest to search on web about paper of "Peter Flach"
Dipartimento di Informatica - Università di Bari
via Orabona, 4 Bari I-70125 - Italy
Tel: +39 0805442299
Fax: +39 0805443269
----- Original Message -----
Sent: Friday, September 02, 2005 12:13 AM
Subject: Wekalist Digest, Vol 31, Issue 2
Send Wekalist mailing list submissions to
To subscribe or unsubscribe via the World Wide Web, visit
or, via email, send a message with subject or body 'help' to
You can reach the person managing the list at
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Wekalist digest..."
1. Re: plese help me...urgent (Ashraf M. Kibriya)
2. what data conform each cluster in K mean algorithm
3. Re: distributing process (Peter Reutemann)
4. programming problem (Stefan Kuhn)
5. Re: programming problem (David)
6. Re: Prediction (Peter Reutemann)
7. Re: Very different results in 2 weka SMO versions? (Eibe Frank)
8. Tertius algorithm (Natália Quirino de Oliveira)
9. Re: Tertius algorithm (Peter Reutemann)
Date: Thu, 01 Sep 2005 12:23:23 +1200
From: "Ashraf M. Kibriya" <amk14(a)cs.waikato.ac.nz>
Subject: Re: [Wekalist] plese help me...urgent
Content-Type: text/plain; charset=us-ascii
Hi, I'm a student of a Bicocca University in
Milan, Italy and is the
first time that I use Weka.
I want to create a text classifier for some documents of my University.
I have got only a database of 142 entry both for train and
test.(multilabel problem with around 100 categories)
I have made the preprocessing phase myself, because my documents are
written in italian and this language is very difficoult to preprocess
with a automatic tool.
Now I have built all file in arff format, and I want to ask if there is
some algorithm that is best for my problem. Now I have used Multiclass
in order to aggregate my file and a neural network, but it take more
than two hour to make a model...It's normal?
Try using Multinomial Naive Bayes or the Complement Naive Bayes, they are
generally more well suited for text classification problems. For higher accuracy, you
might want to try SVM (i.e. SMO classifier in Weka), although it would take longer to
Date: Wed, 31 Aug 2005 21:54:22 -0500
From: Ricardo Farf?n <rijofa(a)gmail.com>
Subject: [Wekalist] what data conform each cluster in K mean algorithm
Content-Type: text/plain; charset="iso-8859-1"
I have a java implementation of clustering analisis, I'm working whit K
but I want to know what data conform each cluster in K mean algorithm
ATT: Ricardo Farf?n