On Friday, November 14, 2003 at 2:05:29 PM, Edgar wrote in the message "Marking mail as Junk with Bayesit filter" <mid:[EMAIL PROTECTED]>:
> Is it wise to mark Spam mail that contains a lot of normal words > in it, as spam when using the Bayesit filter? > You probably seen one of these messages by now. > They look like this: > [SNIP] > This list of words differ and some are more normal then this > example. > When you mark it as spam, won't it upset the statistics? I don't think so. I rarely get e-mail messages containing that random a collection of words that isn't spam. Bayesit filters on combinations and position, not just words. -- Chris <mailto:[EMAIL PROTECTED]> Quoting when replying to this message is good for your karma. Today's Oxymoron: New classic
pgp00000.pgp
Description: PGP signature
________________________________________________ Current version is 2.01.3 | "Using TBUDL" information: http://www.silverstones.com/thebat/TBUDLInfo.html
