Le Wed, Nov 14, 2007 at 11:50:02AM -0600, Manoj Srivastava a écrit :
> The Unsure would be manually inspected, and used to further
> train the filters; as well as any erroneous classification
> (TOE). Periodically, we TUNE (Train Until No Error) the Corpus.
Maybe another heuristic to tri
On Wed, 14 Nov 2007 16:51:42 +0100, Thomas Viehmann <[EMAIL PROTECTED]> said:
> Hi Manoj,
> If you have suggestions how automatic testing can be incorporated into
> the a spam-removal process in a way that is acceptable to the project,
> I'd be very happy to seem them discussed here. However I'm
Hi Manoj,
Manoj Srivastava wrote:
> Hmm. I'll be happy to help automate some of the decision making
> using my Spam classification mechanisms; please look at
>http://www.golden-gryphon.com/software/spam/crm114_accuracy.html
> to see the lower bound on accuracy I get from (mostly) De
On Mon, 12 Nov 2007 00:24:19 +0100, Thomas Viehmann <[EMAIL PROTECTED]> said:
> Hi, if people are interested to spam-removal happen, I am looking for
> help testing how this would work. You need
> - a GPG key that is somewhat close to the Debian keyring,
> - to look at hundreds of messages that
4 matches
Mail list logo