subject-tagging test results

[Greg Louis]

On 20030214 (Fri) at 0939:36 -0500, David Relson wrote:
> At 08:06 AM 2/14/03, Greg Louis wrote:
> 
> >This will be written up later today or tomorrow on
> >http://www.bgl.nu/bogofilter, for those who are interested in the
> >details.

It's there now: http://www.bgl.nu/bogofilter/subjtag.html -- the graph
is worth taking a look at if this test interests you at all.

> Nick is seeing that a Robinson spam_cutoff around 0.34 or 0.38 
> works for him - no false positives.  Given my use of Robinson-Fisher and 
> the variation in my unsures, I can't go below 0.885.  I guess what they say 
> is true - "Different strokes for different folks"

You can't compare Robinson-GM cutoffs with Robinson-Fisher ones.  The
transition from nonspam to spam is far steeper with Fisher.  I'm
currently using 0.98 with Fisher, and that gives a false-positive
result "in the grass," around 0.1% for the difficult work environment
and well below that for my own personal mailbox.  False negatives are
under 5% at work -- looks like I can make that 3.5% with subject
tagging, thanks! -- and around 1.5% at home.

-- 
| G r e g  L o u i s          | gpg public key:      |
|   http://www.bgl.nu/~glouis |   finger greg at bgl.nu |
| Help free our mailboxes. Include                   |
|        http://wecanstopspam.org in your signature. |