Mailman 3 April 2006 - SpamBayes

(no subject)
by stephanie salvignol April 26, 2006

April 26, 2006

gggggggg

1 0

Spambayes hangs when training
by Chris Bulen April 26, 2006

April 26, 2006

Hi, I am a noob, so bear with me. I am attempting to train spambayes, and found that when I directed to my inbox of OE, to train as Ham, it takes a super long time, and my CPU cycles go to 100%, and it never seems to finish. Possibly this is due to large numbers of e-mails with attachements that are video and picture files? I recently archived my inbox, and so have e-mails going back only as far as November 2005. Training on my new spam folder (that has only 75 items right now), works fine. Sincerely, Chris Bulen chris(a)cbulen.com

1 0

Re: [Spambayes] Incremental Training for ham in Outlook Plugin?
by Jesse Pelton April 26, 2006

April 26, 2006

This is one for the training gurus. You can find a discussion of various training approaches on the SpamBayes wiki (http://www.entrian.com/sbwiki/TrainingIdeas). That said, I'll put my oar in. In general, the recommendation of the gurus is along the lines of "don't worry, be happy:" as long as you're getting satisfactory results, just use the training buttons to correct classification errors. The bottom line is the quality of the results you're getting; the suggestion to keep the ham:spam ratio close to 1 is a guideline that seems to help achieve that result. I follow that approach, and when I notice that I'm getting unsatisfactory results over a period of time, I just discard my training database and start over. SpamBayes learns very quickly, so I don't find it worthwhile to try to tune the database over time. Another thing to look at is the threshold scores for possible and certain spam. I've dropped my certain spam threshold somewhat as I've become more confident in my training data (it's now .70). This means fewer possible spam messages that I then train as spam, which reduces the ham:spam imbalance. I'm currently getting good results (>95% correctly classified) with 53 ham and 171 spam trained on. ________________________________ From: spambayes-bounces(a)python.org [mailto:spambayes-bounces@python.org] On Behalf Of Gil Hurlbut Sent: Monday, April 24, 2006 4:35 PM To: spambayes(a)python.org Subject: Re: [Spambayes] Incremental Training for ham in Outlook Plugin? The question addresses the fact that SpamBayes is far better at classifying ham once it is trained than it is in keeping up with classifying new spam. I find it necessary to remove many spam messages until I get to the point where the Manager has far more spam than ham. Until I hear a recommendation differently, I'm going to get back to a balance by moving known ham to my Unsure folder and click on "Recover from Spam" to do the incremental training.

2 1

Problem with zip files
by Gouichoux Steven April 25, 2006

April 25, 2006

Hi, In my organization some problems appear with a combination SpamBayes-Outlook2003-Exchange5.5. In fact, when these users wanted to join a file in a mail, all files were available.. except zip files !! It was a special spambayes problem, because only computers with spam-bayes were touched. Solution found : install a new profile with the same parameters as before. Launch it. Close it. Re-launch the first profile. Zip files are now available to be joined. Crazy, don't you think ? Have you already undergone a problem like this ? Have you got an other solution to go over it ? Thanks a lot. Steven, from France

1 0

Instalation Problem
by Tony FreeSurf April 25, 2006

April 25, 2006

1.. My Outlook Express states it can not find localhost The host 'pop.localhost' could not be found. Please verify that you have entered the server name correctly. Account: 'pop.infonie.fr', Server: 'pop.localhost', Protocol: POP3, Port: 110, Secure(SSL): No, Socket Error: 11001, Error Number: 0x800CCC0D No doubt it is because I have not carried out this Instruction : Setup your mail client: You need to set your mail client to receive mail from localhost rather than your mail server. But Spambayes does not tell me how to carry out this instruction =You need to set your mail client to receive mail from localhost rather than your mail server Best Regards Tony. Tony Arnold

3 2

Re: [Spambayes] Incremental Training for ham in Outlook Plugin?
by Jesse Pelton April 24, 2006

April 24, 2006

I don't think I understand the question. Training SpamBayes incrementally is the preferred approach and designed to be easy. You train on anything that is not classified correctly, either by clicking the Spam or Not Spam buttons as appropriate or by dragging an incorrectly classified message to the folder it belongs in. You take no training action on anything that's correctly classified. So, let's say a message shows up in your inbox that should have been classified as spam. You can either select it and click the Spam button or drag it to your spam folder. If you find ham in your spam or unsure folder, click Not Spam or drag it to your inbox. Ham in your inbox you read; spam in your spam folder you gleefully delete. (That's what I do, anyway.) If that doesn't help, maybe someone else can take a crack at answering. ________________________________ From: spambayes-bounces(a)python.org [mailto:spambayes-bounces@python.org] On Behalf Of Gil Hurlbut Sent: Monday, April 24, 2006 3:44 PM To: spambayes(a)python.org Subject: Re: [Spambayes] Incremental Training for ham in Outlook Plugin? My concern is the ongoing training. Rather than restarting the configuration I want to use new ham to keep up with the new spam I have identified. I'm looking for confirmation that the approach to take to do incremental training is to move filtered ham out of my Inbox and then select "Recover from Spam".

2 1

Re: [Spambayes] spambayes and maildrop (xfilter)
by b52＠entrap.de April 24, 2006

April 24, 2006

Hi, well I solved the problem. In the virtual users home directory was an old .spambayesrc file in a wrong format. The sb_filter.py tried to interpret that file and crashed. I haven't seen that problem by invoking sb_filter.py manually because it was a virtual user that only exists in mysql. The real system user for mail is "vmail". If I invoked the command manually as vmail user, spambayes looked in /home/vmail for a .spambayesrc file which doesn't exist, so it worked. If maildrop invoked that command with xfilter, $HOME was /home/vmail/domain/user where the wrong .spambayesrc existed, so it failed. bye, b52 > > Hi guys, > > maybe someone can help me with this one. I configured postfix to transport > every mail with maildrop and included a maildrop rule to pipe incomming > mails to sb_filter.py > If this happens I get the following error in mail.log: > > Apr 24 03:08:20 hades postfix/pipe[29142]: 6F04D27CA: to=<b52(a)entrap.de>, > relay=maildrop, delay=3, status=deferred (temporary failure. Command > output: Traceback (most recent call last): File "/usr/bin/sb_filter.py", > line 80, in ? from spambayes import hammie, > Options, mboxutils, storage File > "/usr/lib/python2.3/site-packages/spambayes/hammie.py", line 5, in ? > from spambayes import storage File > "/usr/lib/python2.3/site-packages/spambayes/storage.py", line 68, in ? > from spambayes import classifier File "/usr/ > lib/python2.3/site-packages/spambayes/classifier.py", line 73, in ? > from spambayes.Options import options File > "/usr/lib/python2.3/site-packages/spambayes/Options.py", line 1263, in ? > load_options() File "/usr/lib/python2.3/site-packages/spambayes/Option > s.py", line 1195, in load_options options.merge_files(alts) File > "/usr/lib/python2.3/site-packages/spambayes/OptionsClass.py", line 521, in > merge_files self.merge_file(f) File > "/usr/lib/python2.3/site-packages/spam > > Well to tell you the truth I don't understand that message at all. Maybe > some python geeks do.. > My maildrop rule is: > > xfilter "/usr/bin/sb_filter.py -d /home/vmail/bayes.db" > > I tried xfilter with another binary and it works fine. > Thanks for any help > bye, b52 >

1 0

Re: [Spambayes] Incremental Training for ham in Outlook Plugin?
by Jesse Pelton April 24, 2006

April 24, 2006

I assume you mean to train as ham when you move a message from a spam folder to the inbox (or as spam when moving the other direction). If this is the case, whether it happens or not depends on how you configure SpamBayes. In the main Outlook window, click the SpamBayes tab, then SpamBayes Manager..., then click the Training tab. There should be checkboxes that allow you specify whether or not training occurs when you move messages. (I'm pretty sure this is in 1.0.4, but I've been running 1.1a1 for so long that I no longer remember what the differences are.) Training on dragging is mentioned in the "Welcome to SpamBayes" topic in the Outlook plug-in's help system. Perhaps it should be more prominent. The SpamBayes authors welcome patches. ________________________________ From: spambayes-bounces(a)python.org [mailto:spambayes-bounces@python.org] On Behalf Of Gil Hurlbut Sent: Monday, April 24, 2006 12:01 PM To: spambayes(a)python.org Subject: [Spambayes] Incremental Training for ham in Outlook Plugin? My intent is to do incremental training on new ham. I believe it can only be done by a "Recover from Spam". So I expect the procedure is to move the new ham out of the Inbox to do this. Is it taken as so self-evident that it is not mentioned in the documentation?

2 1

Incremental Training for ham in Outlook Plugin?
by Gil Hurlbut April 24, 2006

April 24, 2006

My intent is to do incremental training on new ham. I believe it can only be done by a "Recover from Spam". So I expect the procedure is to move the new ham out of the Inbox to do this. Is it taken as so self-evident that it is not mentioned in the documentation?

1 0

spambayes and maildrop (xfilter)
by b52＠entrap.de April 24, 2006

April 24, 2006

Hi guys, maybe someone can help me with this one. I configured postfix to transport every mail with maildrop and included a maildrop rule to pipe incomming mails to sb_filter.py If this happens I get the following error in mail.log: Apr 24 03:08:20 hades postfix/pipe[29142]: 6F04D27CA: to=<b52(a)entrap.de>, relay=maildrop, delay=3, status=deferred (temporary failure. Command output: Traceback (most recent call last): File "/usr/bin/sb_filter.py", line 80, in ? from spambayes import hammie, Options, mboxutils, storage File "/usr/lib/python2.3/site-packages/spambayes/hammie.py", line 5, in ? from spambayes import storage File "/usr/lib/python2.3/site-packages/spambayes/storage.py", line 68, in ? from spambayes import classifier File "/usr/ lib/python2.3/site-packages/spambayes/classifier.py", line 73, in ? from spambayes.Options import options File "/usr/lib/python2.3/site-packages/spambayes/Options.py", line 1263, in ? load_options() File "/usr/lib/python2.3/site-packages/spambayes/Option s.py", line 1195, in load_options options.merge_files(alts) File "/usr/lib/python2.3/site-packages/spambayes/OptionsClass.py", line 521, in merge_files self.merge_file(f) File "/usr/lib/python2.3/site-packages/spam Well to tell you the truth I don't understand that message at all. Maybe some python geeks do.. My maildrop rule is: xfilter "/usr/bin/sb_filter.py -d /home/vmail/bayes.db" I tried xfilter with another binary and it works fine. Thanks for any help bye, b52

1 0