I am finding that whatever parameters are currently used to decide whether or not 2 constituent records are suspected duplicates is wayyyyy to broad.  Hundreds of suspected dupes come up when I run this process who, by any stretch of the imagination , would not be considered duplicates.  I think that there should be a set-up tool for the user to define what parameters they want to use to determine suspected dupes (i.e. date of birth, gender, the number of letters to go into a first or last name to match, etc...) This would produce a far more manageable and realistic list of suspected dupes.