On Wed, Apr 10, 2013 at 12:49 PM, David Rees <dree...@gmail.com> wrote: > So it appears that the training is completely ineffective as I understand > that once an email as been marked as spam, it should no longer consider the > email for the whitelist. > > If you look at the dspam log or look at the dspam webui, the history page > seems to indicate that the email is indeed being retrained successfully. > > Now, if I take one of these emails, remove the dspam headers and train then > as an inoculation source, after retraining around 5 times the email will be > successfully marked as spam. > > How can I debug this issue further?
Here is something interesting. In a sample email that was marked as Innocent, I took a token from the headers and run it through dspam_dump: $ dspam_dump <user> Subject*nodeposit+package 4592615711378120074 S: 00000 I: 00050 P: 0.0100 Then I retrained the message as an error. $ dspam --class=spam --source=error < <plain-text-spam-file> Then I checked dspam_dump again: $ dspam_dump <user> Subject*nodeposit+package 4592615711378120074 S: 00000 I: 00050 P: 0.0100 Still no SPAM hits! Looking up the token in the database there are no spam_hits marked as expected from dspam_dump. Checking the webui the email has been marked as retrained. Then I remove the DSPAM headers from the message, and innoculate the message once: $ dspam --class=spam --source=inoculation < <plain-text-spam-file> And check dspam_dump again: $ dspam_dump customer-support Subject*nodeposit+package 4592615711378120074 S: 00002 I: 00050 P: 0.0434 Look at that, the spam probability has gone up! So at this point, it's very clear that error-training is not working at all. Any hints before I start digging in to the code? -Dave ------------------------------------------------------------------------------ Precog is a next-generation analytics platform capable of advanced analytics on semi-structured data. The platform includes APIs for building apps and a phenomenal toolset for data science. Developers can use our toolset for easy data analysis & visualization. Get a free account! http://www2.precog.com/precogplatform/slashdotnewsletter _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
