Hello, when dspam 3.9.0-beta4 came out, I updated mine installation to it and it works fine. I'm also trying out "osb" as Tokenizer and PValue.
As a mail got misclassified I got curious how confident dspam was about it and looked at the dspam headers: ,----[ ham ]- | X-DSPAM-Confidence: 1.0000 | X-DSPAM-Improbability: 1 in 98689409 chance of being spam | X-DSPAM-Probability: 0.0000 `---- ,----[ spam ]- | X-DSPAM-Confidence: 1.0000 | X-DSPAM-Improbability: 1 in 98689409 chance of being ham | X-DSPAM-Probability: 1.0000 `---- Looking at some more mails all had the same values. >From an older ham (before the upgrade to beta4): ,----[ old ham ]- | X-DSPAM-Confidence: 0.9916 | X-DSPAM-Improbability: 1 in 11813 chance of being spam | X-DSPAM-Probability: 0.0000 `---- ,----[ old ham ]- | X-DSPAM-Confidence: 0.8381 | X-DSPAM-Improbability: 1 in 519 chance of being spam | X-DSPAM-Probability: 0.0000 `---- The confidence and the improbability value changes for different mails. Is beta4 really such confident about its classification? Michael ------------------------------------------------------------------------------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev _______________________________________________ Dspam-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dspam-user
