HI,
Thanks for the response.

May I know how I can capture the output of the sa trainer ? I using the
follow script to do training,

cd /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/cur
/usr/bin/sa-learn --spam ./*
cp -a /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/cur/* $DIRCOLLECTSPAM
rm -rf /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/cur/*
cd /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/new
/usr/bin/sa-learn --spam ./*
cp -a /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/new/* $DIRCOLLECTSPAM
rm -rf /home/vpopmail/domains/$DOMAIN/$SPAM/Maildir/new/*


I also do the same for the HAM using the same script which section is not
shown here . 


-----Original Message-----
From: John Hardin [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, June 17, 2008 2:54 AM
To: NGSS
Cc: users@spamassassin.apache.org; [EMAIL PROTECTED]
Subject: Re: SA experts needed here - SPAM examples

On Tue, 17 Jun 2008, NGSS wrote:

> I am training SA with around 30-50 manually identified spam (moving spam
> mails to and spam folder created in squirrelmail and crond the sa-train
> command on that folder every hour to train and delete them).

I would suggest hourly is too often (but that may be personal preference), 
and you don't want to delete them. It's a good idea to retain your 
training corpus in case you need to retrain from scratch for some reason.

> However, I found that the learning process is either not right or it is
> rather slow.

What does the learning process report? Are you capturing the output of the 
cron'd sa-learn script?

-- 
  John Hardin KA7OHZ                    http://www.impsec.org/~jhardin/
  [EMAIL PROTECTED]    FALaholic #11174     pgpk -a [EMAIL PROTECTED]
  key: 0xB8732E79 -- 2D8C 34F4 6411 F507 136C  AF76 D822 E6E6 B873 2E79
-----------------------------------------------------------------------
   The world has enough Mouse Clicking System Engineers.
                                                        -- Dave Pooser
-----------------------------------------------------------------------
  2 days until SWMBO's Birthday

Reply via email to