I have encountered a possible bug in the "Subset Matcher"-node regarding the "Maximum mismatches". I have prepared a minimum example to reproduce the behaviour from the field of association anaylsis.
There are 3 sets of items constituting the antecedents in corresponding association rules. When the "Maximum mismatches" are set to 0 the outcome is as expected. When these are set to 1 regarding the fuzzy matching restriction I would expect the LHS-set  not to match the second and 4th purchase which do not contain the item 01. In contrast to this the LHS-set  is not matched against the first 3 purchases. This seems inconsistent to me.
Knime Version 3.1.1
Windows 7, no further details accessible due to shrinked user priviliges