Re: Spamprobe -D option

>>>>> "Brian" == Brian Burton <[email protected]> writes:

Brian> This is exactly what I'd expect from the README.TXT file:

Brian>      Tells SpamProbe to use the database in the specified directory
Brian>      (must be different than the one specified with the -d option) as a
Brian>      shared database from which to draw terms that are not defined in
Brian>      the user's own database.                          ^^^^^^^^^^^^^^
            ^^^^^^^^^^^^^^^^^^^^^^^

Ah, I had read that but the significance of it didn't sink in.

So using -D is probably doing what I intended, but only in the case where a
spam with the significant terms has been seen by the spamtrap before my
personal database sees it and has a chance to possibly mis-classify it.

I'm tempted to suggest an option which makes the shared db take precedence
over the personal db rather than the other way around.  But, in light of your
statement in another post about unexpected results, I have a feeling that the
result won't be so obviuosly "good" as one might think.  For example, there
may be terms common to a lot of these known-spams that are also common to my
"good" mail, so preferring the shared db's value for those terms could produce
more false positives.

Maybe a better option would be to apply these known spams as training to my
personal db.  I'm going to try that and see what happens.

Thanks for the clue.

Mark
-- 
Mark Costlow    | Southwest Cyberport | Fax:   +1-505-232-7975
[email protected] | Web:   www.swcp.com | Voice: +1-505-232-7992

abq-strange.com -- Interesting photos taken in Albuquerque, NM
                   Last post: Iris Door @ 2006-02-01 01:27:05


-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642