Re: Can your bayes do this?

2016-01-21 Thread Christian Laußat
Am 21.01.2016 13:19, schrieb Reindl Harald: no entirely when "urrently, SA's bayes tokens are single words" from https://mail-archives.apache.org/mod_mbox/spamassassin-dev/201211.mbox/%3c509d55a8.30...@gmail.com%3E is still true please review that response below and consider 2/4 word tokes

Re: Can your bayes do this?

2016-01-21 Thread RW
On Thu, 21 Jan 2016 13:19:20 +0100 Reindl Harald wrote: > Am 21.01.2016 um 13:11 schrieb RW: > > On Wed, 20 Jan 2016 22:21:49 -0800 > > Marc Perkel wrote: > > > >> OK - Just to show you this isn't Bayesian - see if you can do this. > >> > >> Here is a list of 5505874 words and phrases used in

Re: Can your bayes do this?

2016-01-21 Thread RW
On Wed, 20 Jan 2016 22:21:49 -0800 Marc Perkel wrote: > OK - Just to show you this isn't Bayesian - see if you can do this. > > Here is a list of 5505874 words and phrases used in the subject line > of HAM and never seen in the subject line of SPAM > >

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Axb
On 01/21/2016 01:25 PM, Robert Chalmers wrote: I’m looking for a way to just dump mail that has the header modified with the * SPAM * assignment. I mean, not have the Client mail reader do it, just have either spamd, or postfix/dovecot dump it. I’m sure I’ve seen something

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 13:34 schrieb Antony Stone: On Thursday 21 January 2016 at 13:31:29, Reindl Harald wrote: On 01/21/2016 01:25 PM, Robert Chalmers wrote: I’m looking for a way to just dump mail that has the header modified with the * SPAM * assignment. I mean, not have the

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Jari Fredriksson
Antony Stone kirjoitti 21.1.2016 14:34: On Thursday 21 January 2016 at 13:31:29, Reindl Harald wrote: > On 01/21/2016 01:25 PM, Robert Chalmers wrote: >> I’m looking for a way to just dump mail that has the header modified >> with the * SPAM * assignment. >> >> I mean, not have

Re: Can your bayes do this?

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 14:17 schrieb RW: On Thu, 21 Jan 2016 13:45:08 +0100 Christian Laußat wrote: Am 21.01.2016 13:19, schrieb Reindl Harald: no entirely when "urrently, SA's bayes tokens are single words" from

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Antony Stone
On Thursday 21 January 2016 at 13:31:29, Reindl Harald wrote: > > On 01/21/2016 01:25 PM, Robert Chalmers wrote: > >> I’m looking for a way to just dump mail that has the header modified > >> with the * SPAM * assignment. > >> > >> I mean, not have the Client mail reader do it,

Re: Can your bayes do this?

2016-01-21 Thread Dianne Skoll
On Thu, 21 Jan 2016 12:11:15 + RW wrote: > "ambulatory care" -> only in ham ... > is that you have discarded the count information. And his assertion is not necessarily true, either. According to our statistics, we've seen "ambulatory care" in 1400 spams, but

Re: Can your bayes do this?

2016-01-21 Thread RW
On Thu, 21 Jan 2016 13:45:08 +0100 Christian Laußat wrote: > Am 21.01.2016 13:19, schrieb Reindl Harald: > > no entirely when "urrently, SA's bayes tokens are single words" from > > https://mail-archives.apache.org/mod_mbox/spamassassin-dev/201211.mbox/%3c509d55a8.30...@gmail.com%3E > > is still

Re: Can your bayes do this?

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 13:11 schrieb RW: On Wed, 20 Jan 2016 22:21:49 -0800 Marc Perkel wrote: OK - Just to show you this isn't Bayesian - see if you can do this. Here is a list of 5505874 words and phrases used in the subject line of HAM and never seen in the subject line of SPAM

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Robert Moskowitz
I use amavis-new to do this: amavisd.conf $log_level = 1; # set the log level to one $sa_tag_level_deflt = -999; # i want to see the headers so change to -99 $sa_tag2_level_deflt = 5.0; # start with 5 $sa_kill_level_deflt = 9; # change to 9 $sa_dsn_cutoff_level = 9; #

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Robert Chalmers
That looks to be just what I want. I now have it running, so will see how it goes. Thanks for that. Much appreciated There are a few other really good options, but this one is nice and compact, no extra scripts. I’m running amavis-new, postfix with postscreen fairly heavily in use, dovecot,

Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Robert Chalmers
I’m looking for a way to just dump mail that has the header modified with the * SPAM * assignment. I mean, not have the Client mail reader do it, just have either spamd, or postfix/dovecot dump it. I’m sure I’ve seen something about doing this, but can’t find it now…. lost in

Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 13:29 schrieb Axb: On 01/21/2016 01:25 PM, Robert Chalmers wrote: I’m looking for a way to just dump mail that has the header modified with the * SPAM * assignment. I mean, not have the Client mail reader do it, just have either spamd, or postfix/dovecot

Re: Can your bayes do this?

2016-01-21 Thread Dianne Skoll
On Wed, 20 Jan 2016 22:21:49 -0800 Marc Perkel wrote: > Here is a list of 5505874 words and phrases used in the subject line > of HAM and never seen in the subject line of SPAM > Here is a list of 3494938 words and phrases used in the subject line > of SPAM and

Re: Can your bayes do this?

2016-01-21 Thread Antony Stone
On Thursday 21 January 2016 at 13:11:15, RW wrote: > On Wed, 20 Jan 2016 22:21:49 -0800 Marc Perkel wrote: > > OK - Just to show you this isn't Bayesian - see if you can do this. > > > > Here is a list of 5505874 words and phrases used in the subject line > > of HAM and never seen in the subject

Re: Can your bayes do this?

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 07:21 schrieb Marc Perkel: OK - Just to show you this isn't Bayesian - see if you can do this. Here is a list of 5505874 words and phrases used in the subject line of HAM and never seen in the subject line of SPAM http://www.junkemailfilter.com/data/subject-ham.txt Here is

Fixed? Re: Looking for a way to dump spam assassin modified mail

2016-01-21 Thread Robert Chalmers
Just to let anyone else know who may be interested. This appears to have solved the last little bit of spam getting through, as well as removing email that had the addition to the Subject line of the SPAM** signal. I haven’t had ANY spam sneak through now since I implemented this.

Re: Can your bayes do this?

2016-01-21 Thread John Hardin
On Thu, 21 Jan 2016, RW wrote: On Thu, 21 Jan 2016 14:31:09 +0100 Christian Laußat wrote: Am 21.01.2016 14:17, schrieb RW: The FNs dropped from 287 to 69, which I'd call a four-fold improvement. The FPs rose from 0 to 1, but that mail was ham quoting a full spam, so arguably it just did a

Re: Can your bayes do this?

2016-01-21 Thread RW
On Thu, 21 Jan 2016 14:31:09 +0100 Christian Laußat wrote: > Am 21.01.2016 14:17, schrieb RW: > > The FNs dropped from 287 to 69, which I'd call a four-fold > > improvement. > > > > The FPs rose from 0 to 1, but that mail was ham quoting a full > > spam, so arguably it just did a better job in

Re: Rule updates are too old - 2016-01-21

2016-01-21 Thread Axb
On 01/21/2016 05:42 PM, John Hardin wrote: On Thu, 21 Jan 2016, dar...@chaosreigns.com wrote: 20160120: Spam or ham is below threshold of 150,000: http://ruleqa.spamassassin.org/?daterev=20160120 20160120: Spam: 131777, Ham: 142710 Oooo, so close! My spam levels are extremely low so I've

Re: Can your bayes do this?

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 20:38 schrieb RW: On Thu, 21 Jan 2016 08:53:10 -0800 (PST) John Hardin wrote: There was an improvement in FP and FN from two tokens. The marginal improvement from three doesn't seem worth it. The improvement from 2 to 3 is more substantial than from 1 to 2 287/160 =

Re: Rule updates are too old - 2016-01-21

2016-01-21 Thread John Hardin
On Thu, 21 Jan 2016, dar...@chaosreigns.com wrote: 20160120: Spam or ham is below threshold of 150,000: http://ruleqa.spamassassin.org/?daterev=20160120 20160120: Spam: 131777, Ham: 142710 Oooo, so close! -- John Hardin KA7OHZhttp://www.impsec.org/~jhardin/

Re: Can your bayes do this?

2016-01-21 Thread Reindl Harald
Am 21.01.2016 um 17:53 schrieb John Hardin: On Thu, 21 Jan 2016, RW wrote: On Thu, 21 Jan 2016 14:31:09 +0100 Christian Laußat wrote: Am 21.01.2016 14:17, schrieb RW: The FNs dropped from 287 to 69, which I'd call a four-fold improvement. The FPs rose from 0 to 1, but that mail was ham

Re: Can your bayes do this?

2016-01-21 Thread John Hardin
On Thu, 21 Jan 2016, RW wrote: On Thu, 21 Jan 2016 08:53:10 -0800 (PST) John Hardin wrote: There was an improvement in FP and FN from two tokens. The marginal improvement from three doesn't seem worth it. The improvement from 2 to 3 is more substantial than from 1 to 2 287/160 = 1.79

Re: Can your bayes do this?

2016-01-21 Thread RW
On Thu, 21 Jan 2016 08:53:10 -0800 (PST) John Hardin wrote: > There was an improvement in FP and FN from two tokens. The marginal > improvement from three doesn't seem worth it. The improvement from 2 to 3 is more substantial than from 1 to 2 287/160 = 1.79 160/69 = 2.3 Whether any of