Re: [Classifier4j-devel] Training and Classifying

[Nick L. <ni...@ma...>]

Are you referring to the Trainer.java & Analyser.java in the 
net.sf.classifier4J.demo package? If so then these are intended as a 
demo only (and a pretty old one at that).

Classifier4J supports training directly via the teachMatch method (see 
http://classifier4j.sourceforge.net/xref/net/sf/classifier4J/bayesian/BayesianClassifier.html#193).

That supports data persistence via the IWordsDataSource interface, 
in-particular the JDBCWordsDataSource 
<../../../../net/sf/classifier4J/bayesian/JDBCWordsDataSource.html> 
implementation of it.

Nick

np...@nt... wrote:

>Hi,
>
>I would like to use Classifier4J to input "good" and "bad" articles; then 
>provide the system with a new article for classification.
>
>Using Trainer.java I can input my articles into a table.  However, how do I continue adding articles to the table?
>The following line in Trainer.java always creates a new table, I would like to continue adding to an existing table:
>
>JDBCWordsDataSource wds = new JDBCWordsDataSource(cm);
>
>I have a similar problem with Anayser.java.  I would like the new article to be 
>classified against the table built from Trainer.java.  However, the 
>setupClassifier method in Analyser.java always creates a new table.
>
>I would very much appreciate any advice.
>
>Regards
>
>-----------------------------------------
>Email provided by http://www.ntlhome.com/
>
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: SourceForge.net Broadband
>Sign-up now for SourceForge Broadband and get the fastest
>6.0/768 connection for only $19.95/mo for the first 3 months!
>http://ads.osdn.com/?ad_id=2562&alloc_id=6184&op=click
>_______________________________________________
>Classifier4j-devel mailing list
>Cla...@li...
>https://lists.sourceforge.net/lists/listinfo/classifier4j-devel
>
>  
>