Hi, I'm still me,
actually I gave a try to the v1.1a1 and I found some
nice statistics. However I would like to have some others reports, mainly
more on "spamming" statistics than on spambayes capabilities.
In particular I think it could be interesting to add:
1 - total spam, good, false positive and false negative messages in
a day for each day of a month
2 - total spam, good, false positive and false negative messages in
a month for each month of a year
3 - total spam, good, false positive and false negative messages in
a year (for all years spent using spambayes)
4 - mean values for the preceding points (not very mandatory because
they could be calculated from the points above)
So, the real interesting points are the first three ones. With them one
could be able to build histograms or generic plot of spam phenomena and its
evolution.
(Just to know, the commercial antispam software called SpamBully already
have the values I reported above, but I noticed that spambayes is better
than spambully and I would like to have this feature also in spambayes).
Thanks for your time.
See you!
Matteo
> -----Messaggio originale-----
> Da: Tony Meyer [mailto:[EMAIL PROTECTED]
> Inviato: mercoledì 29 giugno 2005 3.29
> A: 'Matteo Merlin'; [email protected]
> Oggetto: RE: [Spambayes] Suggestion for SpamBayes: statistics for
> processedmessages
>
> > I'm using spambayes for 6 months now and I found it amazing!
> > Despite of this I've seen that there aren't any statistics for emails
> > processed by spambayes.
>
> Message processing statistics have been vastly improved in 1.1. If you're
> using the Outlook plug-in, then give 1.1a1 a try, and see what you think.
>
> The tracker for this request is:
>
> [ 765924 ] Spam / ham statistics
> <http://sourceforge.net/tracker/index.php?func=detail&aid=765924&group_id=
> 61
> 702&atid=498106>
>
> Feel free to add comments there if there are other statistics that you
> feel
> would be worth adding.
>
> > CVS would be great also because it could be useful when
> > importing data into an application like excel to make
> > statistical calculation or plot graphs showing spam trends...
>
> The statistics are output to the log file (with the Outlook plug-in) when
> Outlook is closed. You could grab them from there and put them in a file
> of
> some type. If you really feel that being able to "export" the statistics
> into a CSV file would be useful, then please either add this to the
> tracker
> above, or open a new one for it. (I could certainly see this being added,
> if there was demand for it).
>
> =Tony.Meyer
>
> --
> Please always include the list (spambayes at python.org) in your replies
> (reply-all), and please don't send me personal mail about SpamBayes.
> http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
_______________________________________________
[email protected]
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html