Without knowing what you are doing with the data, it's difficult suggest the best approach.
if you simply want to know which are not existing from the second set compared to the first set, you can use the reference row filter node for this, choosing exclude as the option.
You could also add a constant value column to the second dataset using constant value node using the label "flag". Then use the cell replacer node to point to this new flag column, using the id columns as the reference columns.
You could also in the joiner node make sure it doesn't remove the duplicate Id column from the settings in the joiner node so you have two Id columns after the join. You could then use the Rule Engine node to use the logic that if one id column is a value and one id column is missing, then put a "flag" it into a new column.