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.";
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re^2: Metric for confidence of complex match
by japhy (Canon) on Oct 27, 2005 at 23:03 UTC |