testng and tuning [wase: katastrophy]
David Relson
relson at osagesoftware.com
Wed Jan 22 17:05:31 CET 2003
At 10:43 AM 1/22/03, Boris 'pi' Piwinger wrote:
>Test is still runnig, but I gotta go, so here you get the
>results I have so far:
>
>Spam:
> 4186 test.spam
>False negatives:
>364
>Ham:
>
>
>So I am not too happy with the false negatives. What
>parameters can I change without having to rebuild the database?
>
>My understanding is that tweaking min_dev and spam_cutoff
>would be OK, right? How about changing the agorithm to robinson?
pi,
If you're really interested in doing some testing, that'd be great. What I
usually do is create a script that sets up the parameters I want and run
it. I might try something like:
dir="test_dir"
for alg in r f ; do
#build wordlists
rm -f $dir/spamlist.db $dir/goodlist.db
bogofilter -$alg -v -d $dir -s < spam.mbx
bogofilter -$alg -v -d $dir -s < spam.mbx
for dev in 0.0 0.1 ; do
echo "min_dev=$dev" > test.cf
formail < test.mbx -s bogofilter -$alg -d $dir -c test.cf
done
done
I'm sure you get the idea. Anyhow, setup the mailboxes, run the script,
pick the best results. There are other parameters which can be tuned for
optimal results with _your_ email. If you're interested, let me know.
David
More information about the Bogofilter
mailing list