Simon Byrnand said:
> > - for spam, must have 3 head hits and 3 body hits
>
> Why ? This seems a bit arbitrary to me. Either we trust the scoring or we
> don't :) What is magic about 3 in particular ?
Yeah, not sure myself. if I recall correctly it gave a stronger
statistical basis to ensur
At 15:20 22/06/03 -0700, Justin Mason wrote:
Matt Kettler said:
> As for disabling the network checks for auto-learning, that makes sense to
> me as well, since the bayes code learns from text tokens, not IPs.
Actually, not quite right, if you're scanning with network tests, it'll
do the auto-lea
At 10:08 22/06/03 -0400, Matt Kettler wrote:
At 08:30 PM 6/21/03 -0400, Gordon Cormack wrote:
Auto-learn and auto-whitelist use different scoring criteria from those
used in spamassassin's spam filtering.
The bayes auto-learning does not use "it's own" scoring mechanism, it uses
scoreset 0. This i
Matt Kettler said:
> As for disabling the network checks for auto-learning, that makes sense to
> me as well, since the bayes code learns from text tokens, not IPs.
Actually, not quite right, if you're scanning with network tests, it'll
do the auto-learn score test with network tests as well.
Gordon Cormack said:
> In supervised mode, positive feedback is exactly what you want.
>
> For the reasons that I've mentioned before, the lack of feedback in the
> current setup causes the system to 'learn' progressively less accurate
> information.
BTW supervised mode is pretty trivial to set
On Sun, Jun 22, 2003 at 10:45:42AM -0400, Gordon Cormack wrote:
> What I have observed is < 0.2% false > positives and < 1.0% false negatives.
I miscomputed, using only the spam count in the denominator. The true
numbers [false / (ham+spam)] are:
false positives: < 0.05% (counted)
false
On Sun, Jun 22, 2003 at 10:08:07AM -0400, Matt Kettler wrote:
> At 08:30 PM 6/21/03 -0400, Gordon Cormack wrote:
> >Auto-learn and auto-whitelist use different scoring criteria from those
> >used in spamassassin's spam filtering.
>
> The bayes auto-learning does not use "it's own" scoring mechanis
At 08:30 PM 6/21/03 -0400, Gordon Cormack wrote:
Auto-learn and auto-whitelist use different scoring criteria from those
used in spamassassin's spam filtering.
The bayes auto-learning does not use "it's own" scoring mechanism, it uses
scoreset 0. This is the score the email would get by the main S
Hi Gordon
> The rationale for spamassassin's behaviour is, I think, the fear that
> in unsupervised mode it will go off track. Perhaps there should be a user
> flag "supervised/unsupervised" that determines whether or not the same
> criteria are used for filtering and learning. In "supervised" m
Auto-learn and auto-whitelist use different scoring criteria from those
used in spamassassin's spam filtering.
IMO, this is a serious mistake. In the long run, it means that the
bayesian and whitelist algorithms will simply reinforce whatever errors
are made by the feature-based classifier.
I ha
10 matches
Mail list logo
