http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5376





------- Additional Comments From [EMAIL PROTECTED]  2007-08-07 10:46 -------
(In reply to comment #10)
> I propose the following new
> measurement. Let's call it the Findlay measurement:
> 
> F(lambda) = 1 / (FN% + FP% * lambda)
> 
> (Actually as defined above it's a function.)
> 
> This is exactly equal to TCR on a balanced (50/50) corpus and doesn't have the
> "undesirable" properties you mentioned.

it's not bad, but it needs work I'm afraid.  For example,
take a look at this:

  fn=3 fp=1.5 f=0.0128205128205128
  fn=18 fp=1.2 f=0.0128205128205128

in other words, if the FP% in this case dropped from 1.5% to 1.2%, that would
be enough to disguise the FN% rate dropping from 3% to *18%*. ;)





------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to