Most of these have such low occurence rates in the corpus that they shouldn't be allowed to vary their scores by the GA (or at least should be much more tightly constrained). I'll take a look at adding such functionality to the code in /masses.
Precicely because they're so infrequently seen though, I wouldn't worry too much about them, and wouldn't not use the new scoreset just because of these few odd scores. C Duncan Findlay wrote: > Questional tests: > > score GAPPY_TEXT -3.667 > spam: 261. nonspam: 112. > > score PORN_8 -5.452 > spam: 3. nonspam: 22. > > score TRACKER_ID -4.899 > spam: 9. nonspam: 9 > > > Ridiculous scores: > > score 25FREEMEGS_URL -4.606 > spam: 6. nonspam: 0. > > score CYBER_FIRE_POWER -4.020 > spam: 10. nonspam: 0. > > score EXCUSE_5 13.447 > spam: 10. nonspam: 0. > > score MONSTERHUT -8.280 > spam: 40. nonspam: 0. > > score ONCE_IN_LIFETIME -4.604 > spam: 83. nonspam: 5. > > > > _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
