On Понедельник, 5 Январь 2004 10:40, Robert Menschel wrote: > Hello Alexander, > > Sunday, January 4, 2004, 7:43:59 PM, you wrote: > > AL> I have written some tests and don't know what is the score set for > these AL> tests. Can some one make the suggestion how to score tests ? > > I have several algorithms I use. Note that my spam threshold is 9.0 -- so > nobody should use my algorithm as-is unless your threshold is also 9.0 > > Most common is: > > ns = number of spam > > nh = number of ham > > t = ns / ( nh + 1 ) > > t1 = int ( t / 100 ) ; r1 = rem ( t / 100 ) > > t2 = int ( t / 10 ) > > if ( t2 = 0 ) then score = 0.$t (stop) > > if ( t1 < 100 ) then score = 1.r1 (stop) > > if ( t1 < 200 ) then score = 2.r1 (stop) > > else score = 3.0 > > Is that the sort of thing you're looking for? > > Note that because of my special situation, I run these tests and apply > these scores strictly local (no network tests) and without Bayes. > Therefore I only need worry about one scoreset.
Yes, these formulas are what I am looking for... ------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id78&alloc_id371&op=click _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
