RE: [Classifier4j-devel] Next Step after Training?

[Nick L. <nl...@es...>]

Yes, that looks fine.

The line "double result = classifier.classify(bayMsgs[i].getSubject());" is
doing the filtering.

Nick


-----Original Message-----
From: Kashif [mailto:ks...@ai...]
Sent: Thursday, 15 July 2004 3:54 PM
To: cla...@li...
Subject: [Classifier4j-devel] Next Step after Training?
Importance: Low


Hi
 
I solved the earlier problem with the "Exception in thread "main"
java.lang.NoClassDefFoundError:
org/apache/commons/logging/" I had to install the latest version of jdk.
 
Thanks Nick for your help with training the filter
 
I used the following code
 
               // For Black List Arrays
                for (int i=0; i<blackListArray.length; i++) {
                   classifier.teachNonMatch(blackListArray[i]);
                }
                
                // For White List Arrays
                for (int i=0; i<whiteListArray.length; i++) {
                   classifier.teachMatch(whiteListArray[i]);
                }
                
                // For BayMsgs Arrays
                // Applying Bayesian Filter on subject
                // BayMsgs is an array of Message Object.
                
                for (int i=0; i<bayMsgs.length; i++) {
                  double result =
classifier.classify(bayMsgs[i].getSubject());
                }
               
 
Does this look OK to you. 
 
Can you please let me know what are the next steps, and what should I do
next to apply Bayesian filter on the subject field only of bayMsgs Array,
which holds the Message Object.
 
Regards
 
And Thanks for our help.