in reply to Metric for confidence of complex match

I'd probably try an additive approach, but it seems any confidence metric based on these data would be pretty arbitrary. If I understand the problem, you could have zeros for all fields of two domains that are actually "linked" or you could have 3s for all fields of two domains that are not linked (proxies for private registration spring to mind.)

-sauoq
"My two cents aren't worth a dime.";
  • Comment on Re: Metric for confidence of complex match

Replies are listed 'Best First'.
Re^2: Metric for confidence of complex match
by japhy (Canon) on Oct 27, 2005 at 23:03 UTC
    Well, there has to be at least one non-zero field, or the link wouldn't have been determined. And as for proxies, I've already removed all those sorts of false positives. I did that for several hours yesterday. I have a fun job.

    Jeff japhy Pinyan, P.L., P.M., P.O.D, X.S.: Perl, regex, and perl hacker
    How can we ever be the sold short or the cheated, we who for every service have long ago been overpaid? ~~ Meister Eckhart