Improved duplicate detection...
- alter logic so that a steady stream of duplicates will continue to be counted as one group, and a new group ONLY started after specified time, rather than multiple groups
- apply split on ":" or "," so minor value changes are ignored