[prev in list] [next in list] [prev in thread] [next in thread] 

List:       spamassassin-users
Subject:    Re: some concernns in collaboratif effort in learning process for
From:       "John D. Hardin" <jhardin () impsec ! org>
Date:       2006-11-30 15:53:32
Message-ID: Pine.LNX.4.10.10611300750390.8084-100000 () gypsy ! impsec ! org
[Download RAW message or body]

On Thu, 30 Nov 2006, Najib Abi Fadel wrote:

> Create a mail account let's say : spamMails@mysite.com Tell
> trusted users to forward the mails they identify as spam to
> "spamMails@mysite.com" . Run a cron job that launch "sa-learn" on
> the mailbox "spamMails@mysite.com", making the spamassassin
> identify them as spam.

Bad idea. Forwarding a message alters that message, in a way which
make it not appropriate for training Bayes.

A better solution is to create mail folders (either shared or
per-user) on the mail server, one for spam and one for ham, and have
the user copy FNs (uncaught spam) to the spam folder and FPs (ham
misclassified as spam) to the ham folder, then train from the folders
on the server.

I have a script that does training like this that you can run from
cron.daily

http://www.impsec.org/~jhardin/antispam/

--
 John Hardin KA7OHZ                    http://www.impsec.org/~jhardin/
 jhardin@impsec.org    FALaholic #11174     pgpk -a jhardin@impsec.org
 key: 0xB8732E79 -- 2D8C 34F4 6411 F507 136C  AF76 D822 E6E6 B873 2E79
-----------------------------------------------------------------------
  ...to announce there must be no criticism of the President or to
  stand by the President right or wrong is not only unpatriotic and
  servile, but is morally treasonous to the American public.
                                          -- Theodore Roosevelt, 1918
-----------------------------------------------------------------------
 15 days until Bill of Rights day

[prev in list] [next in list] [prev in thread] [next in thread]