On Sun, 14 Mar 2010 16:29:57 +0000 Nigel Frankcom <ni...@blue-canoe.com> wrote:
> Note. After you unlearn stuff in one category it is useful to relearn > it in the other - so spam - ham and ham - spam. You don't normally need to unlearn. Bayes keeps track of what it's learned and will learn and/or unlearn as appropriate. That's assuming you haven't deleted the signatures.