Ted Mittelstaedt wrote on Sun, 31 Aug 2014 07:08:11 -0700: > Out of the box the default decision point of 5 is too high anyway.
No. You can always lower it yourself. With the result of more FPs. If you or your users can live with that. Fine. Many can't. > I think the emphasis on avoiding false positives in the stock > (non-Bayes) distribution is far too high. No. This and the "multi-axe" approach are what kept alive SA over this long time and led to wide adoption in businesses as well and made it a de -facto standard for content-based filtering. Kai -- Get your web at Conactive Internet Services: http://www.conactive.com
