[bogofilter-announce] bogofilter-0.13.5 - new current release

[Greg Louis]

On 20030604 (Wed) at 1102:59 +0200, Boris 'pi' Piwinger wrote:

> I did some new test. Training with 20,000 ham and 10,000
> spam and testing with unknown messages:
> 
> algorithm=fisher
> min_dev=0.025
> ham_cutoff = 0.00
> spam_cutoff = 0.60
> robs=0.0001
> 
> Spam:
>    2769 test.spam
> False negatives:
> 155
> Ham:
>    2184 test.ham
> False positives:
> 4
> ------------------------------------------
> robs=0.001
> 
> Spam:
>    2769 test.spam
> False negatives:
> 111
> Ham:
>    2184 test.ham
> False positives:
> 4
> ------------------------------------------
> robs=0.01
> 
> Spam:
>    2769 test.spam
> False negatives:
> 90
> Ham:
>    2184 test.ham
> False positives:
> 5
> ------------------------------------------
> 
> This again shows for my messages, that the new robs is worse.

It shows nothing of the sort.  It shows that you didn't adjust the spam
cutoff, which must be done, because changing s changes the distribution
of scores.  That's all it shows.

When comparing different s values, you need to adjust the spam cutoff
so your fp rate stays the same, then compare fn.  Otherwise the
comparison is meaningless.


-- 
| G r e g  L o u i s          | gpg public key: finger     |
|   http://www.bgl.nu/~glouis |   glouis at consultronics.com |
| http://wecanstopspam.org in signatures fights junk email |