Matt Tolton writes: > By chi2 "combining", does that mean it affects the way that messages are tokenized > and probabilities inputted into the bayes DB, or is it the way that the end score is > calculated from the DB?
It's how the probabilities are combined into 1 overall probability. > Also, are there any downsides to using it? If so, what? It seems to be a little more sensitive to mistraining -- as it's got a tendency for the far ends of the scales, 0.000 and 0.999. --j. ------------------------------------------------------- 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
