CC'd to Debian bug http://bugs.debian.org/554172
Am Dienstag 03 November 2009 schrieb Bill Yerazunis:
> From: Martin Steigerwald <mar...@lichtvoll.de>
>
> No, that was with the old file. And I created a fresh one due to
> that=20 warning and the debconf warning that on some architectures css
> files might= =20
> have to be recreated anyway.
>
> I just wonder, why CRM114 doesn't classify a mail I trained as ham
> as ham=20 in some cases. I have still some mails in unsure directory
> of mail KMail=20 maildir that are still unsure after learning.
> Shouldn't CRM114 get it on=20 the first training - for this exact same
> trained mail?
>
> Yes, it should - unless there's a preponderance of prior experience
> on the phrases or headers of prior data that make it look like
> it is still spam. (at least, for the OSB and OSBF classifiers,
> it is. Some of the others will learn on a single version and
> override prior experience).
>
> But at least the "in use" error went away.
Yes, that one went away.
Maybe its a treshold issue? Those after learning wronly classified mails
are all above 5 but below 10.
mar...@shambhala:~/Mail/.trash.directory/Unsicher/cur> grep CRM114-Status
*
1257260447.22316.TzOey:2,S:X-CRM114-Status: UNSURE ( 8.75 )
1257260448.22316.oyQo1:2,S:X-CRM114-Status: UNSURE ( 9.35 )
1257261491.22316.huxyz:2,S:X-CRM114-Status: UNSURE ( 8.84 )
1257261491.22316.huxyz:2,S:X-CRM114-Status: UNSURE ( 8.14 )
1257280958.22316.OnA0L:2,S:X-CRM114-Status: UNSURE ( 8.98 )
1257280993.22316.Tapxy:2,S:X-CRM114-Status: UNSURE ( 7.99 )
1257280994.22316.czbwZ:2,S:X-CRM114-Status: UNSURE ( 9.18 )
1257281001.22316.Xp889:2,S:X-CRM114-Status: UNSURE ( 8.63 )
1257283921.22316.6uatN:2,S:X-CRM114-Status: UNSURE ( 7.12 )
("Unsicher" is the german word for "unsure")
I have the classifier:
:clf: /osb unique microgroom/
And the tresholds:
# ---- Things rated equal to or better than this are GOOD email
#:good_threshold: /0.01/
#:good_threshold: /5.0/
:good_threshold: /10.0/
#:good_threshold: /20.0/
# ---- Things rated less than or equal to this are SPAM
#:spam_threshold: /-0.01/
:spam_threshold: /-5.0/
#:spam_threshold: /-10.0/
#:spam_threshold: /-20.0/
These are the defaults of /usr/share/crm114/mailfilter.cf! I had a good
treshold of 5 before the upgrade. I wondered about this change already,
but I thought I'd go with the default. I know that I changed some
threshold due to a recommendation here on the mailing list. Let me see,
whether I find that one:
Here it is:
mar...@shambhala:~/.crm114> bzr diff -r 9..10 mailfilter.cf
=== modified file 'mailfilter.cf'
--- mailfilter.cf 2007-09-19 20:11:10 +0000
+++ mailfilter.cf 2008-03-06 19:36:42 +0000
@@ -378,14 +378,14 @@
#
# ---- Things rated equal to or better than this are GOOD email
#:good_threshold: /0.01/
-#:good_threshold: /5.0/
-:good_threshold: /10.0/
+:good_threshold: /5.0/
+#:good_threshold: /10.0/
#:good_threshold: /20.0/
#
# ---- Things rated less than or equal to this are SPAM
#:spam_threshold: /-0.01/
-#:spam_threshold: /-5.0/
-:spam_threshold: /-10.0/
+:spam_threshold: /-5.0/
+#:spam_threshold: /-10.0/
#:spam_threshold: /-20.0/
# ---- mailfilter uses a single threshold and operates symmetrically.
Should I go back to these settings or should I stick with the new defaults
and hope that CRM114 gets it right over time?
Well its just the good treshold that differs from my old settings anyway.
What do you think?
Should the defaults in the default configuration file that comes with the
debian package be changed as well?
Ciao,
--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7
signature.asc
Description: This is a digitally signed message part.
