For Users of ASSP  <[email protected]> writes:
>Is there any way to remove words or word pairs from the Bayesian
>database?
>
>If you look at the below results of a job application email that I 
>received and analyzed, you will see that none of the words really
>should 
>be indicating that it is spam.  As a matter of fact, if an email
>comes 
>through addressed to our online email address 
>info_at_jollyfarmer_dot_com, the probability of that being bad is
>over 
>98% (see the first line of the analysis)


If you want to train the database use the Email Interface - or "View
Maillog Tail".

EmailSpam:
-False-negative Collection (correctedspam)
Spam that got through -- counts double.

EmailHam:
-False-positive Collection (correctednotspam)
Good mail that was listed as spam, count 4x. 

Rebuildspamdb.pl will also remove corresonding pairs  if there is a
matching EmailHam report


------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
_______________________________________________
Assp-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-user

Reply via email to