On Apr 8, 2008, at 7:35 AM, Steve wrote:
There were 9 SPAM messages I re-trained as SPAM (FN? False
negatives?). One re-trained as ham (FP - False positive?). 161
'innocent' so far (TN? Total negatives - not SPAM?). May I ask to
explain the rest of abbreviations and confirm whether my decryption
was right?
TP = Total SPAM
TN = Total INNOCENT
FP = INNOCENT misclassified
FN = SPAM misclassified
SC = SPAM corpusfed
NC = INNOCENT corpusfed
Also try -H
-H Uses multi-line, human-readable output displaying
the fully-
qualified names for each class of totals,
instead of their
abbreviated terms.
