Re: BAYES_99 score

On Wed, 24 Oct 2012, Cathryn Mataga wrote: On 10/24/2012 8:35 AM, Jari Fredriksson wrote: 24.10.2012 18:19, Ned Slider kirjoitti: > I have had very good success running adjusted scores for BAYES rules, > but I am very careful how I train my bayes database. I've disabled > auto-learning and

Re: BAYES_99 score

On 10/24/2012 8:35 AM, Jari Fredriksson wrote: 24.10.2012 18:19, Ned Slider kirjoitti: I have had very good success running adjusted scores for BAYES rules, but I am very careful how I train my bayes database. I've disabled auto-learning and only manually train on hand-checked ham and spam examp

Re: BAYES_99 score

24.10.2012 18:19, Ned Slider kirjoitti: > I have had very good success running adjusted scores for BAYES rules, > but I am very careful how I train my bayes database. I've disabled > auto-learning and only manually train on hand-checked ham and spam > examples. Consequently, I find the extremes (BA

Re: BAYES_99 score

On 22/10/12 19:15, dar...@chaosreigns.com wrote: On 10/22, JP Kelly wrote: Should I set the BAYES_99 score high enough to trigger as spam? I get plenty of spam getting through which does not get caught because BAYES_99 is the only rule which fires and it is not set to score at or above the thr

Re: BAYES_99 score

On Mon, 22 Oct 2012, dar...@chaosreigns.com wrote: On 10/23, Jari Fredriksson wrote: 22.10.2012 21:15, dar...@chaosreigns.com kirjoitti: Huh, ruleqa doesn't track hits to BAYES_99? If it did, against which database it would do that? It would show the hit rates in the corpora of the masschec

Re: BAYES_99 score

On 10/23, Jari Fredriksson wrote: > 22.10.2012 21:15, dar...@chaosreigns.com kirjoitti: > > Huh, ruleqa doesn't track hits to BAYES_99? > If it did, against which database it would do that? It would show the hit rates in the corpora of the masscheck submitters, like everything else. So, the datab

Re: BAYES_99 score

22.10.2012 21:15, dar...@chaosreigns.com kirjoitti: > Huh, ruleqa doesn't track hits to BAYES_99? If it did, against which database it would do that? Just askin... -- "I'm out of options for now. It is something that has gone wrong "in the apt-get region" (can't find a good expression for that

Re: BAYES_99 score

On 10/22, JP Kelly wrote: > Should I set the BAYES_99 score high enough to trigger as spam? > I get plenty of spam getting through which does not get caught because > BAYES_99 is the only rule which fires and it is not set to score at or above > the threshold. You could. Some people only use ba

Re: BAYES_99 score & lint

John Hardin wrote: On Mon, 22 Jun 2009, MySQL Student wrote: meta LOCAL_BAYES_RTF(BAYES_99 && LOCAL_CTYP_RTF) score LOCAL_BAYES_RTF 1.5 describe LOCAL_BAYES_RTF Rule by AS: Probably an Inline RTF spam If the only thing it's complaining about during lint is the zero sc

Re: BAYES_99 score & lint

On Mon, 22 Jun 2009, MySQL Student wrote: meta LOCAL_BAYES_RTF(BAYES_99 && LOCAL_CTYP_RTF) score LOCAL_BAYES_RTF 1.5 describe LOCAL_BAYES_RTF Rule by AS: Probably an Inline RTF spam mimeheader LOCAL_CTYP_RTFContent-Type =~ /^application\/octet-stream.\.rtf/i score

Re: BAYES_99 score & lint

> > > Post your entire scoring block for LOCAL_BAYES_RTF meta LOCAL_BAYES_RTF(BAYES_99 && LOCAL_CTYP_RTF) score LOCAL_BAYES_RTF 1.5 describe LOCAL_BAYES_RTF Rule by AS: Probably an Inline RTF spam mimeheader LOCAL_CTYP_RTFContent-Type =~ /^application\/octet-stream.\

Re: BAYES_99 score & lint

On 22 Jun, 2009, at 12:50 , MySQL Student wrote: Ideas greatly appreciated. Post your entire scoring block for LOCAL_BAYES_RTF -- Updated to be PRCE compatible after 400 years: /(bb|[^b]{2})/

Re: BAYES_99 score & lint

The debug output is saying that the meta rule, LOCAL_BAYES_RTF, has a dependency, BAYES_99, which has a 0 score. In the score line, there are two zero values. ;) It depends what scoreset you're running in. Also, just because 50_scores.cf has something set doesn't mean something later on doesn't c

Re: BAYES_99 on all mail

Do you have Auto Learning enabled? That helps balance the number of ham and spam messages your system learns. My experience has been that the Bayes database does this eventually if it's feed far more spam than ham. If your spam level is very high, you may want to look at both auto-learn as well

Re: bayes_99 matching since sa-update

What's a "sa-learn --dump magic" output look like? # sa-learn --dump magic 0.000 0 3 0 non-token data: bayes db version 0.000 0297 0 non-token data: nspam 0.000 0 982365 0 non-token data: nham 0.000 0

Re: bayes_99 matching since sa-update

Rolf Loudon wrote: > >> What's a >> "sa-learn --dump magic" output look like? > > # sa-learn --dump magic > 0.000 0 3 0 non-token data: bayes db version > 0.000 0297 0 non-token data: nspam > 0.000 0 982365 0 non-token d

Re: bayes_99 matching since sa-update

hi I use sa-update with channels and updates.spamassassin.org. After the latest run today I am getting matches against BAYES_99 (which adds 3.5) to many messages, where they previously triggered virtually no rules at all. This is causing many false positives, to the extent that I've had

Re: bayes_99 matching since sa-update

Rolf Loudon wrote: > hi > > I use sa-update with channels and updates.spamassassin.org. > > After the latest run today I am getting matches against BAYES_99 > (which adds 3.5) to many messages, where they previously triggered > virtually no rules at all. > > This is causing many false positive

Re: BAYES_99 and ham

On Thu, Jul 26, 2007 at 02:25:30PM +0200, martin f. krafft wrote: > also sprach Joe Zitnik <[EMAIL PROTECTED]> [2007.07.26.1340 +0200]: > >Bump your BAYES_99 score. > > What is the basis for determining the BAYES_* scores? There's If you mean the rule scores: http://wiki.apache.org/spamassass

Re: BAYES_99 and ham

also sprach Joe Zitnik <[EMAIL PROTECTED]> [2007.07.26.1340 +0200]: >Bump your BAYES_99 score. Thanks to all who have replied with this suggestion. Of course I am aware of the possibility to raise the scores, but I would not have turned to this list if it would have been sufficient. I agree

Re: BAYES_99 and ham

> On Thu, 26 Jul 2007, Matus UHLAR - fantomas wrote: > > The downside is, that even if user train much, they'll get _much_ > > of spam hitting only 3.5 points with BAYES_99. On 26.07.07 09:30, John D. Hardin wrote: > There's some reluctance to put Poison Pill rules into the default > distribution

Re: BAYES_99 and ham

On Thursday 26 July 2007 13:40, Joe Zitnik wrote: > Bump your BAYES_99 score. And perhaps even define a BAYES_99_9 and/or BAYES_99_99 rule for bayes probabilities over 99.9% and 99.99%, respectively. I use body BAYES_99 eval:check_bayes('0.99', '0.999') body BAYES_999 eval:c

Re: BAYES_99 and ham

On Thu, 26 Jul 2007, Matus UHLAR - fantomas wrote: > The downside is, that even if user train much, they'll get _much_ > of spam hitting only 3.5 points with BAYES_99. There's some reluctance to put Poison Pill rules into the default distribution... :) If you trust your Bayes training, then inc

Re: BAYES_99 and ham

On Thu, 26 Jul 2007, martin f krafft wrote: Hi list, I just had a flood of spam coming through, which SA classified as ham. On closer inspection, it turns out that the only tests triggered for all those mails were HTML_MESSAGE and BAYES_99. HTML messages are commonplace today (unfortunately),

Re: BAYES_99 and ham

> >martin f krafft writes: > >>I just had a flood of spam coming through, which SA classified as > >>ham. On closer inspection, it turns out that the only tests > >>triggered for all those mails were HTML_MESSAGE and BAYES_99. [...] > >>I know how SA scores are computed. I do wonder how that algori

Re: BAYES_99 and ham

Bump your BAYES_99 score. That's the beauty of spamassassin, it is highly customizable. The determinant of whether it is spam or not is the total score versus your threshold. You can change either. I have my BAYES_99 set to 5.0 points. My threshold is 4.0 points. If there is enough about that

Re: BAYES_99 and ham

Justin Mason schrieb: martin f krafft writes: Hi list, I just had a flood of spam coming through, which SA classified as ham. On closer inspection, it turns out that the only tests triggered for all those mails were HTML_MESSAGE and BAYES_99. HTML messages are commonplace today (unfortunately)

Re: BAYES_99 and ham

martin f krafft writes: > Hi list, > > I just had a flood of spam coming through, which SA classified as > ham. On closer inspection, it turns out that the only tests > triggered for all those mails were HTML_MESSAGE and BAYES_99. > > HTML messages are commonplace today (unfortunately), so they

Re: BAYES_99 issue: spamd using info for nobody, not given spamc user

Jason Levine wrote: > > /usr/bin/spamc -u ${user} -e /usr/sbin/sendmail -oi -f ${sender} $ > {recipient} > > Spamd is running daemonized, with the following options: > > spamd --max-children=10 -d -x -q -i 127.0.0.1 -A > 209.10.108.198,204.193.152.163,192.168.1.163,127.0.0.1 > > I'm using MySQL

Re: BAYES_99 triggered on every message

On Tue, 22 May 2007, Duane Hill wrote: On Tue, 22 May 2007, Jari Fredriksson wrote: Even more strange things... sa-learn -u root dump magic returns data for root, while I have not used root ever (at least not in purpose). spamd runs as root unless you specify a username for it to run as. By

Re: BAYES_99 triggered on every message

Duane Hill wrote: > On Tue, 22 May 2007, Jari Fredriksson wrote: > >> Even more strange things... sa-learn -u root dump magic returns data >> for root, while I have not used root ever (at least not in >> purpose). > > spamd runs as root unless you specify a username for it to run as. By > defa

Re: BAYES_99 triggered on every message

On Tue, 22 May 2007, Jari Fredriksson wrote: Even more strange things... sa-learn -u root dump magic returns data for root, while I have not used root ever (at least not in purpose). spamd runs as root unless you specify a username for it to run as. By default, if no username was used, spam

Re: BAYES_99 triggered on every message

Even more strange things... sa-learn -u root dump magic returns data for root, while I have not used root ever (at least not in purpose). sa-learn -u root -clear cleared this, and NOW spamc -u spam does NOT trigger BAYES_99 any more! It does not trigger any BAYES tests, so it looks it maybe

Re: BAYES_99 triggered on every message

Duane Hill wrote: > > There is one way to find out what username(s) is/are recorded in > MySQL. If you have access to the server running MySQL: > > Get into MySQL by: > > mysql --user=username --password=password name_of_sa_db > > Then: > > select username from userpref; > > This will lis

Re: BAYES_99 triggered on every message

On Sat, 19 May 2007, Jari Fredriksson wrote: Matt Kettler wrote: Jari Fredriksson wrote: b) I have cleaned the database with sa-learn --clean (Still BAYES_99 while the bayes should be off!) What's that do? Did you mean --clear? Yup. Are you sure you did that --clear while logged in as

Re: BAYES_99 triggered on every message

Matt Kettler wrote: > Jari Fredriksson wrote: b) I have cleaned the database with sa-learn --clean (Still BAYES_99 while the bayes should be off!) >>> What's that do? Did you mean --clear? >>> >> >> Yup. >> >> > Are you sure you did that --clear while logged in as the righ

Re: BAYES_99 triggered on every message

Jari Fredriksson wrote: >>> >>> b) I have cleaned the database with sa-learn --clean (Still BAYES_99 >>> while the bayes should be off!) >>> >> What's that do? Did you mean --clear? >> > > Yup. > > Are you sure you did that --clear while logged in as the right user? ie: the same user

Re: BAYES_99 triggered on every message

Matt Kettler wrote: > Jari Fredriksson wrote: >> >> SpamAssassin version 3.1.8 assembled via cpan >> >> Every message gets BAYES_99, even when >> >> a) the message has no body >> >> b) I have cleaned the database with sa-learn --clean (Still BAYES_99 >> while the bayes should be off!) > What's

Re: BAYES_99 triggered on every message

Jari Fredriksson wrote: > > SpamAssassin version 3.1.8 assembled via cpan > > Every message gets BAYES_99, even when > > a) the message has no body > > b) I have cleaned the database with sa-learn --clean (Still BAYES_99 > while the bayes should be off!) What's that do? Did you mean --clear?

Re: BAYES_99 triggered on every message

Craig Carriere wrote: Perhaps a dumb comment on my part, but have you tried to delete the table entries from the mySQL database and are you sure you are using the SA user? Doesn't sa-learn --clean only clear the Berkeley dbs and you appear to state that you are using mySQL. Best --clean creat

Re: BAYES_99 triggered on every message

Perhaps a dumb comment on my part, but have you tried to delete the table entries from the mySQL database and are you sure you are using the SA user? Doesn't sa-learn --clean only clear the Berkeley dbs and you appear to state that you are using mySQL. Best Jari Fredriksson wrote: > > SpamAssas

Re: BAYES_99 triggered on every message

On Wed, May 16, 2007 at 06:38:12PM +0300, Jari Fredriksson wrote: > Every message gets BAYES_99, even when > > a) the message has no body Bayes uses the header as well. > b) I have cleaned the database with sa-learn --clean (Still BAYES_99 while > the bayes should be off!) Then you're not doin

Re: BAYES_99 makes lots of false-positive

Matt Kettler wrote: In sa 2.6x or older, yes.. in sa 3.0.0 or higher, no. First, phrases isn't quite accurate.. bayes stores tokens, and most of the tokens are simply words, not phrases. In SA 3.0.0 or higher the text tokens themselves are not stored, only the SHA1 hash of them is stored.

Re: BAYES_99 makes lots of false-positive

Joshua, C.S. Chen wrote: > Hello folks, > My users speak Chinese. I found that spamassassin seems not working well > about chinese chset (utf8 or big5) on the bayes issue. Many normal mails > (almost) get BAYES_99 score although the real spam also get BAYES_99. It > looks like foreign language like

Re: BAYES_99 makes lots of false-positive

On Thu, Jul 13, 2006 at 03:17:05PM +0800, Joshua, C.S. Chen wrote: > Hello folks, > My users speak Chinese. I found that spamassassin seems not working well > about chinese chset (utf8 or big5) on the bayes issue. Many normal mails > (almost) get BAYES_99 score although the real spam also get BAYES

RE: Bayes_99

ww.dankohn.com/> <tel:+1-650-327-2600> -Original Message- From: List [mailto:[EMAIL PROTECTED] Sent: Wednesday, April 06, 2005 04:57 To: SpamAssassin list Subject: Re: Bayes_99 > List wrote: > >> I did a spamassassin -D --lint and notice an error:- >> >> de

Re: Bayes_99

List wrote: I did a spamassassin -D --lint and notice an error:- debug: bayes corpus size: nspam = 230, nham = 397 debug: tokenize: header tokens for *F = "U*ignore D*compiling.spamassassin.taint.org D*spamassassin.taint.org D*taint.org D*org" debug: tokenize: header tokens for *m = " 1112781784

Re: Bayes_99

List wrote: I did a spamassassin -D --lint and notice an error:- debug: bayes corpus size: nspam = 230, nham = 397 debug: tokenize: header tokens for *F = "U*ignore D*compiling.spamassassin.taint.org D*spamassassin.taint.org D*taint.org D*org" debug: tokenize: header tokens for *m = " 1112781784

Re: Bayes_99

Your Bayes is corrupted. rm .spamassassin/bayes* will wipe out your current database and let you start again, which will require 200 spams and 200 hams. Be careful in your training! I had followed your instruction. Now I am getting BAYES_50, BAYES_60 and BAYES_99. The number of BAYES_99 has redu

RE: Bayes_99

Your Bayes is corrupted. rm .spamassassin/bayes* will wipe out your current database and let you start again, which will require 200 spams and 200 hams. Be careful in your training! - dan -- Dan Kohn -Orig

Re: BAYES_99 = 1.9?

At 08:37 PM 1/18/2005, [EMAIL PROTECTED] wrote: BTW, it looks like bayes_90 has been deprecated. When I run a lint on my local.cf, I get: warning: score set for non-existent rule BAYES_90 Yes, several of the old ranges in 2.64 no longer exist.. For 3.x they changed the ranging a bit, creating som

Re: BAYES_99 = 1.9?

On Mon, 17 Jan 2005, Thomas Arend wrote: > With network test enabled bayes scores lower. This is a problem when the > network test don't fire when the spammer uses a new server. Therefore I have > raised the bayes scores for bayes_99. I seldom get bayes_90 so I didn't raise > the scores for bayes_

Re: BAYES_99 = 1.9?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Am Montag, 17. Januar 2005 19:56 schrieb John Fleming: > > The Bayes_9x scores were a bit low for our needs. So we bumped them > > up. That is one of the great things I like about SA. You can customize > > it to meet your needs. > > > > Shane >

Re: BAYES_99 = 1.9?

> The Bayes_9x scores were a bit low for our needs. So we bumped them > up. That is one of the great things I like about SA. You can customize > it to meet your needs. > > Shane My part of the thread was that the scores don't consistently go -up- with increased percentage - see previous. t

Re: BAYES_99 = 1.9?

Hey John, The Bayes_9x scores were a bit low for our needs. So we bumped them up. That is one of the great things I like about SA. You can customize it to meet your needs. Shane - Original Message - From: "John Fleming" <[EMAIL PROTECTED]>Subject: Re: BAYES_99 =

Re: BAYES_99 = 1.9?

> 1.9 BAYES_99 BODY: Bayesian spam probability is 99 to 100% [score: 1.] And what's the rationale for 95% scoring lower than 80%?? Sorry if this has been previously mentioned. score BAYES_00 0 0 -1.665 -2.599 score BAYES_05 0 0 -0.925 -0.