[Spambayes] Feedback from an ecstatic user
Kenny Pitt
kennypitt at hotmail.com
Mon Jul 19 19:40:55 CEST 2004
Tony Meyer wrote:
>> Total emails trained: Spam: 13126 Ham: 2479 More statistics:
>>
>> SpamBayes has processed 9009 messages - 1566 (17%) good,
>> 7131 (79%) spam and 312 (3%) unsure.
>> 648 messages were manually classified as good (0 were false
>> positives). 2835 messages were manually classified as spam (1
>> was a false negative). 9 unsure messages were manually
>> identified as good, and 147 as spam.
>>
>> (I presume the difference in numbers e.g. (13126 vs. 7131) is due to
>> the fact that I have gone through several different different
>> versions)
>
> Most probably, yes. Older versions didn't keep around the same
> information that newer ones do, so that's probably what's caused
> this.
Differences in the counts can also be caused by manual training. 13126 is
the number of spams *trained* while 7131 is the number of spams *processed*
by SpamBayes, so they're counting different things. If you've ever used the
"Train on a message, mbox file or dbx file" option to train on additional
spam messages, those messages will be included in the trained count but not
in the processed count.
--
Kenny Pitt
More information about the Spambayes
mailing list