Re: [SAtalk] unlearning a lot of ham
jeff covey wrote: on Wed, Jun 18, 2003 at 10:50:22AM -0400%, Ross Vandegrift said: ross if you retrain Bayes with the messages in question apparently, this thread has wandered far enough that my original message has been forgotten. it's here: http://marc.theaimsgroup.com/?l=spamassassin-talkm=105586966928433w=2 here's the question again: jeff 75 spam messages have been recorded as ham. i don't have jeff the original messages, so i can't use sa-learn to unlearn jeff them. how can i keep the spam i've given the filter but jeff wipe out all the ham, so i can start training ham from jeff scratch? AFAIK you can't. Either wipe out your Bayes database and begin again, or wait for the token bias to cure itself - which, given time, it will, of course. Depending on how much spam and non-spam you have in the database and the daily volume of mail, the bias for 75 false spam messages shouldn't take too long to decrease. Best, Tony -- Tony Earnshaw Working to get a life http://j-walk.com/blog/docs/conference.htm http://www.billy.demon.nl Mail: [EMAIL PROTECTED] --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
Re: [SAtalk] unlearning a lot of ham
On Tue, Jun 17, 2003 at 09:43:55PM -0400, jeff covey wrote: while the side issues are interesting, does anyone have answers for the actual questions that started this thread? :) LOL, yes - if you retrain Bayes with the messages in question and specifically mention that it's spam, they should be removed from the ham database. -- Ross Vandegrift [EMAIL PROTECTED] A Pope has a Water Cannon. It is a Water Cannon. He fires Holy-Water from it.It is a Holy-Water Cannon. He Blesses it. It is a Holy Holy-Water Cannon. He Blesses the Hell out of it. It is a Wholly Holy Holy-Water Cannon. He has it pierced.It is a Holey Wholly Holy Holy-Water Cannon. He makes it official. It is a Canon Holey Wholly Holy Holy-Water Cannon. Batman and Robin arrive. He shoots them. --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
Re: [SAtalk] unlearning a lot of ham
on Wed, Jun 18, 2003 at 10:50:22AM -0400%, Ross Vandegrift said: ross if you retrain Bayes with the messages in question apparently, this thread has wandered far enough that my original message has been forgotten. it's here: http://marc.theaimsgroup.com/?l=spamassassin-talkm=105586966928433w=2 here's the question again: jeff 75 spam messages have been recorded as ham. i don't have jeff the original messages, so i can't use sa-learn to unlearn jeff them. how can i keep the spam i've given the filter but jeff wipe out all the ham, so i can start training ham from jeff scratch? thanks, -- ++ | jeff covey [EMAIL PROTECTED] http://pobox.com/~jeff.covey/ 410-869-8088 | || | I expect you will send me a great hep.-- freshmeat.net contributor | ++ pgp0.pgp Description: PGP signature
Re: [SAtalk] unlearning a lot of ham
On Tue, Jun 17, 2003 at 12:53:57PM -0400, jeff covey wrote: hello, all. # train spamassassin's bayesian filter macro index S | sa-learn --single --spam -D macro pager S | sa-learn --single --spam -D # report spam to razor macro index Z | spamassassin -r -D macro pager Z | spamassassin -r -D From what I understand 'spamassassin -r' updates the bayesian database as well as reporting to Razor, so the first 2 macros are redundant. I don't think that this has anything to do with your problem, though. Regards: Jim Ford -- Spam poison - don't use! --- [EMAIL PROTECTED] --- --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
Re: [SAtalk] unlearning a lot of ham
On Tue, Jun 17, 2003 at 02:26:14PM -0400, jeff covey wrote: on Tue, Jun 17, 2003 at 07:06:52PM +0100%, Jim Ford said: jim so the first 2 macros are redundant. i don't understand what you're trying to say. one learns spam (for spam i don't want to report) and one learns and reports it. you'll need to explain what's redundant. I didn't realise you'd want to learn spam but not report it! Why would you want to do this - I always report spam that slips through to Razor? Regards: Jim Ford -- Spam poison - don't use! --- [EMAIL PROTECTED] --- --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
Re: [SAtalk] unlearning a lot of ham
on Tue, Jun 17, 2003 at 09:22:18PM +0100%, Jim Ford said: jim I didn't realise you'd want to learn spam but not report it! jim Why would you want to do this - I always report spam that jim slips through to Razor? i assume that other people could use reports of generic spam i've received, but that it won't help them to have a checksum of a message which is customized with my name, email address, etc. it's not going to help me to know that you received a spam which says dear jim ford, here's a hot stock tip just for you...; i doubt that i'll ever receive a copy with your name. am i not understanding something about how this works? -- ++ | jeff covey [EMAIL PROTECTED] http://pobox.com/~jeff.covey/ 410-869-8088 | || | You have reached the freshmeat.net staff, and we don't understand you. | | -- Ray Shaw | ++ pgp0.pgp Description: PGP signature
Re: [SAtalk] unlearning a lot of ham
On Tue, Jun 17, 2003 at 06:44:06PM -0400, jeff covey wrote: i assume that other people could use reports of generic spam i've received, but that it won't help them to have a checksum of a message which is customized with my name, email address, etc. it's not going to help me to know that you received a spam which says dear jim ford, here's a hot stock tip just for you...; i doubt that i'll ever receive a copy with your name. am i not understanding something about how this works? Interestng point! I understand DCC uses checksums, so the above point should apply. I wonder if Razor would also be affected? Regards: Jim Ford -- Spam poison - don't use! --- [EMAIL PROTECTED] --- --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
RE: [SAtalk] unlearning a lot of ham
?!? Both razor and DCC are supposed to be largely insensitive to this e.g. From http://www.rhyolite.com/anti-spam/dcc/FAQ.html Do the fuzzy checksums ignore personalizations? Yes, they ignore many so called personalizations. Balam -Original Message- From: Jim Ford [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 17, 2003 5:08 PM To: [EMAIL PROTECTED] Subject: Re: [SAtalk] unlearning a lot of ham On Tue, Jun 17, 2003 at 06:44:06PM -0400, jeff covey wrote: i assume that other people could use reports of generic spam i've received, but that it won't help them to have a checksum of a message which is customized Interestng point! I understand DCC uses checksums, so the above point should apply. I wonder if Razor would also be affected? --- This SF.Net email is sponsored by: INetU Attention Web Developers Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php ___ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
Re: [SAtalk] unlearning a lot of ham
while the side issues are interesting, does anyone have answers for the actual questions that started this thread? :) most importantly: can i remove all the (bogus) ham data in my bayes database without removing any of the (valid) spam data i've collected? as for reporting leading to learning as ham, i was directed to: http://bugzilla.spamassassin.org/show_bug.cgi?id=2027 and replied to theo privately with: http://jeffcovey.net/tmp/sa/ i hope that helps. thanks, -- ++ | jeff covey [EMAIL PROTECTED] http://pobox.com/~jeff.covey/ 410-869-8088 | || |Hello there, I looked up on your website and I saw many good product there | |and I think I interest on your product : -do you have in stock Casio| |Calculators CASIO GRAPH CALC? -CASIO FX7400G ? -- freshmeat.net contributor| ++ pgp0.pgp Description: PGP signature
Re: [SAtalk] unlearning a lot of ham
on Tue, Jun 17, 2003 at 05:30:52PM -0700%, Balam Willemsen said: q Do the fuzzy checksums ignore personalizations? a Yes, they ignore many so called personalizations. err. thanks, but that's not really enough information to tell me anything. what is considered a personalization? does that just mean it will ignore email addresses and dear $name,, or will it also deal well with something like: http://www.DirectPartners.net/rotation.asp?RefId=78868OptinId=117181814 ? well, i don't really need to know the details. if it's true that razor, pyzor, etc. somehow do the right thing(tm) with all spam, no matter how it's tailored, just knowing that would be enough for me, and i'll be happy to report it all. i just don't want to abuse reporting services by submitting information no one else can use. thanks, -- ++ | jeff covey [EMAIL PROTECTED] http://pobox.com/~jeff.covey/ 410-869-8088 | || | Hi! I think that this site, would be more photos about them. I hope for | | answer!!!... ...If you could send some photo... I\264ll thank\264s| |-- freshmeat.net contributor| ++ pgp0.pgp Description: PGP signature