Am 2013-02-27 04:19, schrieb Chad M Stewart: > I originally had setup my system and trained a "user" with spam/ham, > then put everyone in a global group. Earlier this month I removed > everyone from the global group and reset my account from a dspam > perspective. Since the best accuracy is supposed to be from learning > my behavior. I'm now past the training period but the accuracy seems > low. Have I done something incorrect? > It does not look like you have made anything incorrect. 91.5% accuracy slightly after the training phase is not bad. What tokenizer are you using?
> > > > TP True Positives: 217 > TN True Negatives: 2896 > FP False Positives: 1 > FN False Negatives: 288 > SC Spam Corpusfed: 0 > NC Nonspam Corpusfed: 0 > TL Training Left: 0 > SHR Spam Hit Rate 42.97% > HSR Ham Strike Rate: 0.03% > PPV Positive predictive value: 99.54% > OCA Overall Accuracy: 91.50% > > > > # dspam_admin agg pref c...@balius.com > trainingMode=TOE > spamAction=deliver > spamSubject=[SPAM] > statisticalSedation=5 > enableBNR=on > enableWhitelist=on > signatureLocation=message > tagSpam=off > tagNonspam=off > showFactors=off > optOut=off > whitelistThreshold=10 > storeFragments=off > processorBias=on > trainPristine=off > notifications=on > ignoreRBLLookups=off > > > Thank you, > Chad > > -- Kind Regards from Switzerland, Stevan Bajić ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_feb _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
