<html>
<head>
<style>
P
{
margin:0px;
padding:0px
}
body
{
FONT-SIZE: 10pt;
FONT-FAMILY:Tahoma
}
</style>
</head>
<body>Hi Greg,<BR>
<BR>
What should i do in case that the mail box on the exchange? Can i simply copy those emails to a folder on a linux machine and run sa-learn?<BR>
<BR>
Thanks,<BR>
<BR>
Yossi<BR><BR><BR><BR><BR>
<HR id=stopSpelling>
<BR>
> Date: Fri, 9 Feb 2007 16:17:10 -0500<BR>> From: gborders@jlewiscooper.com<BR>> To: mailscanner@lists.mailscanner.info<BR>> Subject: Re: Scanning for Spam<BR>> <BR>> DAve wrote:<BR>> > am.lists wrote:<BR>> >>> I also meant to point out that your Bayes was only hitting 50% which <BR>> >>> add<BR>> >>> nothing to the score. Start feeding these emails into the Bayes<BR>> >>> learning system, and it will start to match these emails.<BR>> >><BR>> >> So, I guess that's the question.<BR>> >><BR>> >> Is there a way to make Bayes learn this when it's not currently <BR>> >> tagged as spam?<BR>> ><BR>> > There are many many ways to make that happen depending on your <BR>> > network, your policies, your users. We simply keep messages that we <BR>> > feel are spam and once a day I ftp the mailbox up to my MS servers and <BR>> > run sa-learn on them.<BR>> ><BR>> > I can run through our Postmaster mailboxes and gleen 50 a day easily <BR>> > (we have no spam scanning on postmaster or abuse). A simplistic <BR>> > approach, but so far seems to be working. We may have to get serious <BR>> > about it farther down the road.<BR>> ><BR>> > DAve<BR>> ><BR>> >><BR>> >> PS: I did get the additional SARE rules added to my rdj config. I hope<BR>> >> the admins over there don't ban me since I've downloaded some files<BR>> >> than 1x/day today. :-/<BR>> >><BR>> >> Angelo<BR>> ><BR>> ><BR>> Here I made a "spam" mail box, and created SYM links to it for users <BR>> within their mail folders. As folks find spam that leak thru, they drag <BR>> it over the "universal" spam box, and then I feed them via a daily cron <BR>> job to sa-learn script which reads them, and then deletes them.<BR>> <BR>> Also handy is the Mailwatch interface, where I can run searches/reports, <BR>> and then checkbox the bad ones for instant SA learning. Works great!<BR>> <BR>> Greg.<BR>> <BR>> <BR>> <BR>> <BR>> --<BR>> This transmission may contain information that is privileged, confidential<BR>> and/or exempt from disclosure under applicable law. If you are not the<BR>> intended recipient, you are hereby notified that any disclosure, copying,<BR>> distribution, or use of the information contained herein (including any<BR>> reliance thereon) is STRICTLY PROHIBITED. If you received this transmission<BR>> in error, please immediately contact the sender and destroy the material in<BR>> its entirety, whether in electronic or hard copy format. Thank you.<BR>> <BR>> -- <BR>> This message has been scanned for viruses and<BR>> dangerous content by MailScanner, and is<BR>> believed to be clean.<BR>> <BR>> -- <BR>> MailScanner mailing list<BR>> mailscanner@lists.mailscanner.info<BR>> http://lists.mailscanner.info/mailman/listinfo/mailscanner<BR>> <BR>> Before posting, read http://wiki.mailscanner.info/posting<BR>> <BR>> Support MailScanner development - buy the book off the website! <BR><BR><br /><hr />Live Search: New search found <a href='http://get.live.com/search/overview ' target='_new'>Try it!</a></body>
</html>