Is bogofilter Bayesian?

[Greg Louis]

On 20040210 (Tue) at 1337:31 +0100, Boris 'pi' Piwinger wrote:

> > I prefer, however, not to (mis)lead people into thinking it doesn't
> > matter what you train with, nor how many times you do so; I would be
> > sorry to see the warning removed.
> 
> Well, the argument isn't correct.

Says you :)

> Very interesting here is what Liudvikas Bukys added about
> AdaBoost
> (http://kiew.cs.uni-dortmund.de:8001/mlnet/instances/81d91e8d-dc15ed23e9)
> which maybe gives a theoretical background for what I found
> just by trying and from very unscientific ideas.

This wouldn't be worth replying to (it was discussed several months
ago) but that people have short memories.  Adaboost and related
techniques will very effectively train for recognition of a member of a
population, at the expense of capability to recognize something
_similar_ to the members of a population.  Whether a particular
training method proves effective or overspecific needs to be determined
empirically.

-- 
| G r e g  L o u i s         | gpg public key: 0x400B1AA86D9E3E64 |
|  http://www.bgl.nu/~glouis |   (on my website or any keyserver) |
|  http://wecanstopspam.org in signatures helps fight junk email. |