On 2016-06-08 18:53, Joe Quinn wrote:
Usually you don't want to be autolearning at all, and only train with messages that have been reviewed by a human. It's very easy for a Bayes DB to spiral out of control after even just one or two wrong results.
will setting this to -50 make it autolearn spam but disable nearly autolearn ham ?
learning manuel make just a harder work for bayes in the way it generate more digest to compare ?
