Re: [Assp-test] Feature request

[Paul S. <pa...@it...>]

Thanks Thomas !

On Wed, 2010-06-30 at 07:25 +0200, Thomas Eckardt wrote:
> Paul,
> 
> for now set 'baysConf' to 0 .
> 
> This is a general logical mistake inside ASSP V2 - if the Bayes check has 
> switched on the '[lowconfidence]' flag, there is no reason to pass the 
> message, if the last spam detection flag does not belong to the Bayes 
> check - for example 'messagescore'. This will be changed in 2.0.2_1.1.14.
> 
> Thomas
> 
> 
> 
> Von:    Paul Schutte <pa...@it...>
> An:     ASSP development mailing list <ass...@li...>
> Datum:  29.06.2010 17:05
> Betreff:        Re: [Assp-test] Feature request
> 
> 
> 
> 
> Hi.
> 
> I see messages in the maillog.txt that looks like this:
> 2010-06-29 16:52:07 23121-19670 [Worker_15]
> [MessageLimit][lowconfidence] 67.195.8.63 <fa...@at...> to:
> pa...@it... [spam found] and possibly passing because of low
> confidence, otherwise blocked (MessageScore 89, limit 80) [Thank you for
> using Western Union]
> 
> It seems like we could use a switch like:
> "Block messages with Low confidence but High Score"
> 
> This is just a wish. 
> It is already doing a great job and thank you very much for that.
> 
> Regards
> Paul
> 
> On Sat, 2010-06-26 at 20:34 +0200, Paul Schutte wrote:
> > Hello again.
> > 
> > Fritz, thank you for your advice about the (switchTestToScoring)
> > setting.
> > 
> > However, I stil see messages above (PenaltyMessageLimit) being delivered
> > to the users. These messages are tagged as spam (score 40-80 should be
> > tagged). Messages with scores above 80 should be blocked. I would like
> > to block these no matter how the score arrived at 81 and above.
> > 
> > The messages that "slip through" always seems to involve the "bayesian
> > uncertainty factor".
> > 
> > I have none of the testmodes enabled. It automatically goes into
> > testmode for scores from 40-80 and because of the Bayesian uncertainty.
> > This IS what I want, but I would want to drop messages with scores above
> > 80.
> > 
> > Here are the ASSP headers of 2 of these messages:
> > 
> > X-Assp-Version: 2.0.2(1.1.13) on spamcontrol.up.ac.za
> > X-Assp-Message/IP-Score: 25 (DNSBL: neutral, 109.187.25.5 listed in
> >  ix.dnsbl.manitu.net)
> > X-Assp-DNSBL: neutral, 109.187.25.5 listed in
> >  (ix.dnsbl.manitu.net<-127.0.0.2; )
> > X-Assp-Message/IP-Score: 15 (PTR invalid 'h109-187-25-5.dyn.bashtel.ru')
> > X-Assp-Message/IP-Score: 20 (URIBL: neutral,  listed in
> >  studynextrumultisurblorg)
> > X-Assp-URIBL: neutral,  listed in studynextrumultisurblorg
> > X-Assp-Message/IP-Score: 39 (Bayesian Probability: 1.0000)
> > X-Assp-Tag: Bayesian
> > X-Assp-ID: spamcontrol.up.ac.za 99307-19376
> > X-Assp-Redlisted: Yes (bounces are not collected)
> > X-Assp-Spam: YES (Probably)
> > X-Assp-Block: NO (testmode)
> > X-Assp-Spam-Reason: Bayesian
> > X-Assp-Message-Totalscore: 99
> > X-Assp-Tag: MessageLimit
> > X-Assp-Spam-Reason: MessageScore 99, limit 80
> > 
> > 
> > X-Assp-Version: 2.0.2(1.1.13) on spamcontrol.up.ac.za
> > X-Assp-Message/IP-Score: 20 (No Spoofing Allowed 'ne...@cc...')
> > X-Assp-Message/IP-Score: 30 (PTR invalid
> >  '109-93-66-75.dynamic.isp.telekom.rs')
> > X-Assp-Message/IP-Score: 39 (Bayesian Probability: 1.0000)
> > X-Assp-Tag: Bayesian
> > X-Assp-ID: spamcontrol.up.ac.za 52413-00543
> > X-Assp-Spam: YES (Probably)
> > X-Assp-Block: NO (testmode)
> > X-Assp-Spam-Reason: Bayesian
> > X-Assp-Message-Totalscore: 89
> > X-Assp-Tag: MessageLimit
> > X-Assp-Spam-Reason: MessageScore 89, limit 80
> > 
> > 
> > Regards
> > Paul
> > 
> > On Tue, 2010-06-08 at 21:12 +0200, Fritz Borgstedt wrote:
> > > ASSP development mailing list <ass...@li...>
> > > schreibt:
> > > >The easiest to explain this is that had I turned Bayesian off, the
> > > >score
> > > >would have been 110 and the message blocked.
> > > >Now with Bayesian on, the score is 129.5,but message only gets tagged
> > > >and passed on because of Bayesian uncertainty.
> > > 
> > > 
> > > What version are you running?
> > > 
> > > Is ->Switch Testmode to Message Scoring (switchTestToScoring) set?
> > > 
> > > 
> ------------------------------------------------------------------------------
> > > ThinkGeek and WIRED's GeekDad team up for the Ultimate 
> > > GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
> > > lucky parental unit.  See the prize list and enter to win: 
> > > http://p.sf.net/sfu/thinkgeek-promo
> > > _______________________________________________
> > > Assp-test mailing list
> > > Ass...@li...
> > > https://lists.sourceforge.net/lists/listinfo/assp-test
> > 
> > 
> > 
> ------------------------------------------------------------------------------
> > This SF.net email is sponsored by Sprint
> > What will you do first with EVO, the first 4G phone?
> > Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> > _______________________________________________
> > Assp-test mailing list
> > Ass...@li...
> > https://lists.sourceforge.net/lists/listinfo/assp-test
> 
> 
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Assp-test mailing list
> Ass...@li...
> https://lists.sourceforge.net/lists/listinfo/assp-test
> 
> 
> 
> 
> DISCLAIMER:
> *******************************************************
> This email and any files transmitted with it may be confidential, legally 
> privileged and protected in law and are intended solely for the use of the 
> 
> individual to whom it is addressed.
> This email was multiple times scanned for viruses. There should be no 
> known virus in this email!
> *******************************************************
> 
> 
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________ Assp-test mailing list Ass...@li... https://lists.sourceforge.net/lists/listinfo/assp-test