Robert Bartlett wrote: > Are there any "starter" ham/spam emails I can use? I thought I saw one, but > it wasn't for a MySQL database. Im using Bayes, site wide, in mysql. I hate > to feed it emails I think is ham or spam.
Personally I violently detest "starter" bayes databases, and feel that in the general case they are a patently stupid idea. In my opinion they defeat most of the purpose of using bayes in the first place. Bayes's strength is custom adaptation to YOUR site's email patterns. A "starter" database will reflect someone else's email, not yours. Also, due to SA's extensive tokenization of message headers, half the tokens in such a database will be meaningless at your site. I'd suggest doing feeding on your own real mail. If you're concerned about implications for how your users might regard the mail, leave it out. Even better, solicit a couple of your users for a handful of mail messages to include in the training. Pretty much the only time you want to use a "Starter" database is if you don't have enough data to make your own starter training.
