Thread: [Dspam-user] dspam and small spam mails
Brought to you by:
paulcockings,
sbajic
From: ML m. <mln...@ya...> - 2013-11-25 22:53:30
|
Hello, I have the case where there is a mail account which receives a lot of spam mails which does not get marked as spam by dspam. The pattern of this mails looks like this: a few random words (3-7) with even sometime words which do not exist or ar written weirdly (v1ägRÄ) followed by a link which always changes URL. It looks like it is somehow impossible for dspam to tag them as spam. Do you have any recommendations or special settings for these kind of mails? or is it simply not possible? Best regards ML |
From: Steve F. <ste...@fs...> - 2013-11-26 10:19:47
|
On 25/11/13 22:53, ML mail wrote: > Hello, > > I have the case where there is a mail account which receives a lot of > spam mails which does not get marked as spam by dspam. The pattern of > this mails looks like this: a few random words (3-7) with even sometime > words which do not exist or ar written weirdly (v1ägRÄ) followed by a > link which always changes URL. > > It looks like it is somehow impossible for dspam to tag them as spam. Do > you have any recommendations or special settings for these kind of > mails? or is it simply not possible? > Have you been training DSPAM on these messages as spam? Regards, Steve. |
From: ML m. <mln...@ya...> - 2013-11-26 10:49:40
|
Hi Steve, Yes I have been training dspam with these messages by moving them into my Spam folder which automatically retrains dspam. I did this more than 100 of these kind of spam mails but without any success. Best, ML On Tuesday, November 26, 2013 11:25 AM, Steve Freegard <ste...@fs...> wrote: On 25/11/13 22:53, ML mail wrote: > Hello, > > I have the case where there is a mail account which receives a lot of > spam mails which does not get marked as spam by dspam. The pattern of > this mails looks like this: a few random words (3-7) with even sometime > words which do not exist or ar written weirdly (v1ägRÄ) followed by a > link which always changes URL. > > It looks like it is somehow impossible for dspam to tag them as spam. Do > you have any recommendations or special settings for these kind of > mails? or is it simply not possible? > Have you been training DSPAM on these messages as spam? Regards, Steve. ------------------------------------------------------------------------------ Shape the Mobile Experience: Free Subscription Software experts and developers: Be at the forefront of tech innovation. Intel(R) Software Adrenaline delivers strategic insight and game-changing conversations that shape the rapidly evolving mobile landscape. Sign up now. http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk _______________________________________________ Dspam-user mailing list Dsp...@li... https://lists.sourceforge.net/lists/listinfo/dspam-user |
From: Steve F. <ste...@fs...> - 2013-11-26 12:34:53
|
On 26/11/13 10:49, ML mail wrote: > Hi Steve, > > Yes I have been training dspam with these messages by moving them into > my Spam folder which automatically retrains dspam. I did this more than > 100 of these kind of spam mails but without any success. > Which tokenizer are you using? Are you sure the training is working? What probability is DSPAM giving these messages? Regards, Steve. |
From: ML m. <mln...@ya...> - 2013-11-26 14:06:51
|
I am using the teft as TrainingMode. When I retrain dspam with that specific mail by moving it into the Spam folder I can see a new entry being written into the system.log file, here below would be an example: 1385462923 M <None Specified> 4,52946e75120681586112679 <None Specified> 0.121289 my...@my...d Retrained So I can only assume here that the training is working. Or else how can I be 100% sure that the training is working? And here would be the relevant DSPAM headers as taken from one of these spam mails: X-DSPAM-Result: Innocent X-DSPAM-Confidence: 0.6000 X-DSPAM-Improbability: 1 in 151 chance of being spam X-DSPAM-Probability: 0.0010 Let me know if you would need any other details. I would be happy to provide you with these. Best,ML On Tuesday, November 26, 2013 1:40 PM, Steve Freegard <ste...@fs...> wrote: On 26/11/13 10:49, ML mail wrote: > Hi Steve, > > Yes I have been training dspam with these messages by moving them into > my Spam folder which automatically retrains dspam. I did this more than > 100 of these kind of spam mails but without any success. > Which tokenizer are you using? Are you sure the training is working? What probability is DSPAM giving these messages? Regards, Steve. ------------------------------------------------------------------------------ Shape the Mobile Experience: Free Subscription Software experts and developers: Be at the forefront of tech innovation. Intel(R) Software Adrenaline delivers strategic insight and game-changing conversations that shape the rapidly evolving mobile landscape. Sign up now. http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk _______________________________________________ Dspam-user mailing list Dsp...@li... https://lists.sourceforge.net/lists/listinfo/dspam-user |
From: Steve F. <ste...@fs...> - 2013-11-27 13:18:51
|
On 26/11/13 14:06, ML mail wrote: > I am using the teft as TrainingMode. When I retrain dspam with that > specific mail by moving it into the Spam folder I can see a new entry > being written into the system.log file, here below would be an example: > > 1385462923 M <None Specified> 4,52946e75120681586112679 > <None Specified> 0.121289 my...@my...d Retrained > > So I can only assume here that the training is working. Or else how can > I be 100% sure that the training is working? > > And here would be the relevant DSPAM headers as taken from one of these > spam mails: > > X-DSPAM-Result: Innocent > X-DSPAM-Confidence: 0.6000 > X-DSPAM-Improbability: 1 in 151 chance of being spam > X-DSPAM-Probability: 0.0010 > > Let me know if you would need any other details. I would be happy to > provide you with these. > Personally I wouldn't use TEFT; I'm using TOE here. Which tokenizer are you using? OSB? Do you have any training left? e.g. run dspam_stats -H -S my...@my...d Regards, Steve. |
From: ML m. <mln...@ya...> - 2013-11-27 16:53:28
|
I will try switching to TOE. Regarding your other questions we are using OSB as tokenizer yes. Below are the stats for the account I am testing with: TP True Positives: 298 TN True Negatives: 373 FP False Positives: 0 FN False Negatives: 80 SC Spam Corpusfed: 0 NC Nonspam Corpusfed: 0 TL Training Left: 2127 SHR Spam Hit Rate 78.84% HSR Ham Strike Rate: 0.00% PPV Positive predictive value: 100.00% OCA Overall Accuracy: 89.35% btw I would be running this mail server for hosting many domains. Regards, ML On Wednesday, November 27, 2013 2:24 PM, Steve Freegard <ste...@fs...> wrote: On 26/11/13 14:06, ML mail wrote: > I am using the teft as TrainingMode. When I retrain dspam with that > specific mail by moving it into the Spam folder I can see a new entry > being written into the system.log file, here below would be an example: > > 1385462923 M <None Specified> 4,52946e75120681586112679 > <None Specified> 0.121289 my...@my...d Retrained > > So I can only assume here that the training is working. Or else how can > I be 100% sure that the training is working? > > And here would be the relevant DSPAM headers as taken from one of these > spam mails: > > X-DSPAM-Result: Innocent > X-DSPAM-Confidence: 0.6000 > X-DSPAM-Improbability: 1 in 151 chance of being spam > X-DSPAM-Probability: 0.0010 > > Let me know if you would need any other details. I would be happy to > provide you with these. > Personally I wouldn't use TEFT; I'm using TOE here. Which tokenizer are you using? OSB? Do you have any training left? e.g. run dspam_stats -H -S my...@my...d Regards, Steve. ------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk _______________________________________________ Dspam-user mailing list Dsp...@li... https://lists.sourceforge.net/lists/listinfo/dspam-user |