On Wed, 21 Jan 2004 13:48:30 -0500
Ben Hanson <[EMAIL PROTECTED]> wrote:
> I have had Spamassassin running for long enough to have BAYES kick in.
> I realized the other day when I checked out all my rulesets with
> spamassassin -D --lint, I was getting a BAYES_0 test result. I never
> saw suc
I have had Spamassassin running for long enough to have BAYES kick in.
I realized the other day when I checked out all my rulesets with
spamassassin -D --lint, I was getting a BAYES_0 test result. I never
saw such a thing from my email. I figured out that I needed to
specifically run spamc a
At 03:01 PM 11/28/2003, Greg Webster wrote:
No Bayes scores are listed in the header above, so I assume the 'real'
score is 9.0. Is there something else that may be affecting the 'real'
score?
Just to prove the point, I decided to add it all up using both scoreset 3
and scoreset 1.
score BIZ_TLD
On Fri, 2003-11-28 at 12:55, Matt Kettler wrote:
> I'm not sure why the rounded score you have shows up as 9.0.. have you
> altered some scores a bit?
8.652 should round to 8.7, not to 9.0, correct? I haven't altered any of
those scores...the only score changes I've made were my own additions
(pr
At 03:01 PM 11/28/2003, Greg Webster wrote:
No Bayes scores are listed in the header above, so I assume the 'real'
score is 9.0. Is there something else that may be affecting the 'real'
score?
Yes.. scoresets.. disabling bayes does not simply remove BAYES_* from the
list of rules.. it also changes
Heya!
On Fri, 2003-11-28 at 07:27, Matt Kettler wrote:
> At 01:48 PM 11/27/03 -0800, Greg Webster wrote:
> >An example header:
> >X-Spam-Status: Yes, hits=9.0 required=5.0 tests=BIZ_TLD,CLICK_BELOW,
> > DATE_IN_PAST_96_XX,HTML_50_60,HTML_LINK_CLICK_HERE,HTML_MESSAGE,
> > HTTP_EXCESSIV
At 01:48 PM 11/27/03 -0800, Greg Webster wrote:
An example header:
X-Spam-Status: Yes, hits=9.0 required=5.0 tests=BIZ_TLD,CLICK_BELOW,
DATE_IN_PAST_96_XX,HTML_50_60,HTML_LINK_CLICK_HERE,HTML_MESSAGE,
HTTP_EXCESSIVE_ESCAPES,MIME_HTML_NO_CHARSET,MIME_HTML_ONLY,
PENIS_ENLARGE,WLS
On Thu, 2003-11-27 at 14:30, [EMAIL PROTECTED] wrote:
> > 'grep bayes /etc/mail/spamassassin/local/cf`:
>
> The filename is "/etc/mail/spamassassin/local.cf". And don't forget to
> restart spamd/amavisd-ne after any change there.
Just a typo, and I definitely restarted spamd.
Thanks,
Greg
--
I believe I've set things correctly, but SA doesn't seem to be
autolearning:
An example header:
X-Spam-Status: Yes, hits=9.0 required=5.0 tests=BIZ_TLD,CLICK_BELOW,
DATE_IN_PAST_96_XX,HTML_50_60,HTML_LINK_CLICK_HERE,HTML_MESSAGE,
HTTP_EXCESSIVE_ESCAPES,MIME_HTML_NO_CHARSET,MIME_HTML_O
On Monday 17 November 2003 01:29 pm, Bart Schaefer wrote:
> On Mon, 17 Nov 2003, Brian Godette wrote:
> > On Monday 17 November 2003 11:22 am, Bart Schaefer wrote:
> > > :0 c : $HOME/.sa$LOCKEXT
> > >
> > > * ^X-Spam-Checker-Version:.*-$SAVERSION
> > > * ^X-Spam-Status: Yes.*autolearn=ham
> > >
> >
On Mon, 17 Nov 2003, Brian Godette wrote:
> On Monday 17 November 2003 11:22 am, Bart Schaefer wrote:
> > :0 c : $HOME/.sa$LOCKEXT
> > * ^X-Spam-Checker-Version:.*-$SAVERSION
> > * ^X-Spam-Status: Yes.*autolearn=ham
> > | sa-learn --forget
>
> There's one problem with the above and that's spam t
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Michael V. Sokolov writes:
>
>I've noticed message with such headers:
>
>X-Spam-Flag: YES
>X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on antispam
>X-Spam-Level: **
>X-Spam-Status: Yes, hits=6.0 required=5.0 tests=BAYES_99 aut
On Monday 17 November 2003 11:22 am, Bart Schaefer wrote:
> My solution is this little bit of procmail:
> :0 c : $HOME/.sa$LOCKEXT
>
> * ^X-Spam-Checker-Version:.*-$SAVERSION
> * ^X-Spam-Status: Yes.*autolearn=ham
>
> | sa-learn --forget
>
> Replace $SAVERSION with whatever appears in your version_
On Mon, 17 Nov 2003, Brian Godette wrote:
> > X-Spam-Status: Yes, hits=6.0 required=5.0 tests=BAYES_99 autolearn=ham
> > version=2.60
>
> In my opinion this is an annoying flaw in 2.6 as any spam that doesn't
> hit any of the standard rules is learned as ham, if the default
> auto-learn t
On Monday 17 November 2003 08:16 am, Michael V. Sokolov wrote:
> I've noticed message with such headers:
>
> X-Spam-Flag: YES
> X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on
> antispam X-Spam-Level: **
> X-Spam-Status: Yes, hits=6.0 required=5.0 tests=BAYES_99 autolearn=ha
I've noticed message with such headers:
X-Spam-Flag: YES
X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on antispam
X-Spam-Level: **
X-Spam-Status: Yes, hits=6.0 required=5.0 tests=BAYES_99 autolearn=ham
version=2.60
X-Spam-Report:
* 6.0 BAYES_99 BODY: Bayes
I've noticed message with such headers:
X-Spam-Flag: YES
X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on antispam
X-Spam-Level: **
X-Spam-Status: Yes, hits=6.0 required=5.0 tests=BAYES_99 autolearn=ham
version=2.60
X-Spam-Report:
* 6.0 BAYES_99 BODY: Bayes
Hi all, (this might be quite long)
I use RH9, spamass-milter, SA 2.60, sendmail and fetchmail as a front
end server for my Win2000/Exchange network. The principal reason for
placing a linux box at the edge of the network was to filter all
incoming mail with Spamassassin. It works very well (thou
At 05:11 PM 10/7/03 +0200, Jim Knuth wrote:
Hallo SA-List,
why does it give permanently autolearn=no?
This is an header from the sa-list.
I think this must autolearn=ham
--snip
X-Spam-Status: No, hits=-1.4 required=3.5 tests=BAYES_00,DE_PLING_03,
DE_PLING_04,MANY_EXCLAMATIONS,PLING_PLING
Hallo SA-List,
why does it give permanently autolearn=no?
This is an header from the sa-list.
I think this must autolearn=ham
--snip
X-Spam-Status: No, hits=-1.4 required=3.5 tests=BAYES_00,DE_PLING_03,
DE_PLING_04,MANY_EXCLAMATIONS,PLING_PLING autolearn=no
version=2.60
--snap
On Tue, Sep 23, 2003 at 02:16:28PM -0500, Chris Cook wrote:
> Question... for some reason, even if I explicitly enable autolearn in
> the config file it always shows autolearn=no in the headers of a scanned
> email, is there something I'm missing?
no means it wasn't, not that it won't. Run with -
Thanks! Solved my question!
Theo Van Dinter wrote:
>
> On Tue, Sep 23, 2003 at 02:16:28PM -0500, Chris Cook wrote:
> > Question... for some reason, even if I explicitly enable autolearn in
> > the config file it always shows autolearn=no in the headers of a scanned
> > email, is there somethin
Question... for some reason, even if I explicitly enable autolearn in
the config file it always shows autolearn=no in the headers of a scanned
email, is there something I'm missing?
Running SpamAssassin version 2.6 on FreeBSD with sendmail and
spamass-milter, I've checked to make sure the the rul
i notied autolearn=ham does that mean if a spam passes as ham it will learn it has
ham, how can i disable this autolearn feature and what does it really mean, i did not
find anything in the doc
---
This sf.net email is sponsored by:ThinkGee
Hello List:
My setup is as follows:
/---\/\/--\/-\
| Virus Scanner | -> | qmail + SA | -> | Exchange | -> | Outlook |
\---/\/\--/\-/
Although there are some clients using POP3/IMAP, most of th
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
I read about some problems in conjunction with Spamassassin's autolearn
mechanism.
My questions:
Is it momentary reasonable to use the autolearn mechanism?
Will the data format of the db files be shifted to another format in
spamassassin 2.7 (for example)?
And are there problems in conjunction wi
36 matches
Mail list logo
