Because without a few hundred messages, it would be completely and utterly useless? It would be like meeting an airline pilot who was 5'7" tall and had a scar on his left cheek and wore his hat backwards. Bayes would think that scars on left cheeks were as reliable an indicator of airline-pilotness as was an airline uniform. All statistics are based upon having a statistically significant sample, and to tell the truth, a corpus of 200 hams / 200 spams is severely stretching it.
http://www.paulgraham.com/spam.html > -----Original Message----- > From: Ron Gilbert [mailto:[EMAIL PROTECTED] > Sent: Tuesday, September 02, 2003 10:54 AM > To: [EMAIL PROTECTED] > Subject: RE: [SAtalk] bayes feedback > > > > >>I read here that bayes is only turned on after > >>it learned from at least 200 spams AND 200 hams. > >>That number could be more. It only starts to be > >>efficient after you got say 1000 of both. > > Can someone explain to me why SA won't start using bayes > until it's seen > several hundred messages? > ------------------------------------------------------- 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