better Bayesian bogofilter
Boris 'pi' Piwinger
3.14 at logic.univie.ac.at
Tue Aug 12 14:20:27 CEST 2003
Greg Louis wrote:
>> > Is it correct to assume, that this would come with virtually
>> > no cost, since the more complicated calculation is not done
>> > very often in each run of bogofilter?
>>
>> I think that's safe to assume.
>>
> By "more complicated calculation" is meant "equation #5", right?
Yes.
> Yeah, just once per token ;)
So only a couple of hundred more divisions per mail.
> On the other hand, if you train with every
> message or randomly select errors-and-unsures to keep the ratio right,
> you get to use equation #4, which saves 3 divisions per token over what
> we do now.
I guess, most people (including me) don't care about the
ratio in the database.
pi
More information about the Bogofilter
mailing list