[Spambayes] RE: How low can you go?
Robert K. Coe
bob at 1776.com
Wed Dec 17 17:26:47 EST 2003
> From: Seth Goodman [mailto:nobody at spamcop.net]
> Sent: Wednesday, December 17, 2003 2:01 PM
> To: spambayes at python.org; spambayes-dev at python.org
> Subject: RE: [Spambayes] How low can you go?
> Does CRM114 use the number of trained ham and trained spam *messages*
> as variables in its probability calculation? If not, then you wouldn't
> expect that deleting infrequently used tokens would do much damage.
> AFAIK, SpamBayes uses the trained message counts in the probability
> calculation and those becomes inaccurate if you delete individual tokens.
If you delete, say, 5% of the tokens in the database, reduce the message count by 5% as well.
More information about the Spambayes