Currently, if a user was to automatch a specific munge process, in the match validation screen it shows a specific number of rows now marked as matched, and then it also shows the remainder as UNMATCH.
The user would assume that the match validation screen should mark the remainder as NO-MATCH.
The user who assumes that is not clear on the meanings of those statuses.
MATCH: The pair of records definitely represents the same thing. One is the master and the other is the duplicate.
UNMATCH: The pair of records has similar field values, but has not been flagged as having a master/duplicate relationship
NOMATCH: The pair of records has similar field values, but the two records have been confirmed not to represent the same thing.
When you auto-match, a master record is randomly chosen within each "cluster" of matches, then all other records in that cluster are considered duplicate. The presence of any NOMATCH records in the cluster would be a contradiction, since the same pair of records would simultaneously represent the same thing and be definitely not the same thing. This is impossible unless you are Schrödinger's cat.