View Full Version : tet mining with weka

06-22-2008, 05:13 AM
I'm a italian student at unical university.
I need to make text mining with weka.
can you tell me where i can find a tutorial for text mining in weka?

06-22-2008, 05:21 PM

Here is a wiki page that will help you get started with the conversion of text documents into ARFF files for Weka classifiers to learn from.


Beyond this, any general tutorial on approaches to text mining will help you proceed. Weka has a number of classifiers that are known to work well for text categorization: support vector machines, multinomial naive bayes, k nearest neighbors, and, coming in Weka 3.5.8 very soon: bayesian logistic regression and discriminative multinomial naive Bayes.


07-01-2008, 03:51 PM
For proper text mining (of unstructured data, look at the uimaj project.. It was developed by IBM and is now an apache incubator project.

I've downloaded and tried itit, and it seems quite promising, but very "academic" at this point in time.