Page 1 of 1

Unduplicate Independent vs Transitive

Posted: Mon Oct 08, 2012 2:42 am
by mfecdsx
I read the document about these two methods but still I don't understand the difference.

Can someone explain the difference with some example?

Thank you.

Posted: Tue Oct 16, 2012 4:09 pm
by rjdickson
The way I understand it (and I could be off base...):

* Unduplicate Independent finds 'master' records within groups (blocks). It then looks for records to group. For example, if A is the selected master record, and:
A matches C
B matches C
A does not match B
Then QualityStage would match A and C. Remember that A was the master, so everything is being compared to A. B would have to be brought into the group in another pass. Furthermore, the fact that B matched C is not remembered outside this pass.

* Unduplicate Transitive matches every record to every record in the group (block). There is no concept of a 'master' in this case. So:
A matches C
B matches C
A does not match B
Since A matches C and C matches B, therefore A, B, and C are all grouped together.

I hope this helps!