On Mon, Apr 04, 2011 at 02:22:05PM +0300, Ibrahim Harrani wrote: > Hi, > > I know that I should train dspam with error when a message previously > misclassified by dspam. > What happen if I training all email with --source=corpus instead of error? > Does this will make a very important negatif effect for the classification? > > Thanks. >
You need "error" to have DSPAM correct the stats, instead of just add a new bit of information to use. "error" should allow DSPAM to be more sensitive to, uhh, errors. :) Regards, Ken ------------------------------------------------------------------------------ Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
