On Wednesday, August 27, 2003, at 09:10 PM, Simon Byrnand wrote:
By the way, in case anyone thinks I'm slamming them, I wasn't, I'm sure their system is probably a competant implemenation of a Bayesian Classifier, however the claims that were made over the phone and in an email forwarded to me were definately exagerated...
Unless you do a LOT of training, I don't see how a purely Bayesian classifier can perform as well as something as multifacited as SpamAssassin. (I wonder how well a purely Bayesian classifier does on the HTML image only spams for example)
I use SpamAssassin with the built in (matching) rules and Bayes only -- no RBLs, Pyzor, Razor, DCC or other network based rules. Most days I get 95% or more spam caught just with those. This is with 2.55 btw.
Chad
------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
