scripts
Tom Allison
tallison at tacocat.net
Fri Mar 12 13:04:29 CET 2004
FYI
I came up with this little script to use for training/testing to
exhaustion. Works fairly well on my email archives of ham/spam.
Designed for maildir format mail (courier-imap).
I have another script that juggles mail around my spam/ham into these
two folders on a cron basis and then I can use this to test/adjust my
word database to ensure that any new training doesn't push any old
archives into the wrong category, which has happened. New spam/ham
training can cause old spam/ham archives to start reporting as 'Unsure'.
This fixes the problem.
>99%, but I'm always striving for more!
--------------------------------------------------------
#!/bin/sh
cd /home/tallison/Maildir/.training.ham/cur/
for F in `ls`; do
bogofilter -v < $F | grep -v No && bogofilter -nv < $F &&
bogofilter -v < $F
;
done
cd /home/tallison/Maildir/.training.spam/cur/
for F in `ls`; do
bogofilter -v < $F | grep -v Yes && bogofilter -sv < $F &&
bogofilter -v < $
F;
done
More information about the Bogofilter
mailing list