several independent groups

[rderidder-lda]

say my scenario is that over years, there are Events created, and each Event has participants associated.
Problem is that sometimes participants are duplicated.

I do NOT want to match between events.. but I want to get all the matches WITHIN each event.
How to best do this, knowing there are maybe 1 million events, each with a 4 or 5 participants.

I thought of using the 'EXACT' on the event (event id), and use text/shortstring matching on the participant names...and then let it run on the whole set...
However it seems that even with EXACT, sometimes it will stray away and match across..
Is there a better way or is there a way to be 100% sure it never matches across event ids? Make the Exact field100% required to be identical, and the other fields use the thresholds, etc.

maybe i should use Categorical.. but the categories would be the million event ids.. doesn't seem like its the right thing to do

Thanks
R