GETTING.STARTED (rev 2)
Boris 'pi' Piwinger
3.14 at piology.org
Wed Oct 27 12:07:03 CEST 2004
David Relson said:
>> > 6. Tuning bogofilter
>> > --------------------
>> >
>> > Once you've use bogofilter for a while, you may wish to optimize
>> > its classification parameters. The bogotune utility uses your
>> > wordlist and additional ham and spam messages to check a large
>> > variety of possible parameter values and find what'll work best
>> > for your environment. For more info, read the bogotune man page
>> > and file bogofilter-tuning.HOWTO.html.
>>
>> This then applies only to some training methods.
>
> Eh?? Bogotune uses the wordist, and the ham and spam corpora you
> specify, and then does a rather exhaustive scan of possible scoring
> parameters to find what gives the best results. As you know, bogotune
> has minimum requirements for number of messages registered in
> wordlist.db and minimum numbers of messages for the ham and spam corpora
> used in the tuning process.
Right, so it is not usable for pure train-on-error approaches.
> The only effect that training method has is whether or not enough
> messages are present in the wordlist for bogotune to work successfully.
There is still this unsolved question if tuning works for train-on-error
(or it becomes somehow circular). My guess is that it works, but is
practically impossible.
> cc: bogofilter mailing list
Thanks, it was too late at night, so I forgot.
pi
More information about the Bogofilter
mailing list