Consider two classifiers C-1 and C-2 each partitioning a collection A
of n objects into classes based on respective sets of features. In a n
umber in pattern recognition and knowledge engineering applications, t
he following problems arise: given the two partitions generated by C-1
and C-2, compute new partitions by considering (i) the features commo
n to both classifiers, and (ii) the features of either of the classifi
ers. We show that both problems can be solved in Theta(n log n) time u
sing very simple algorithms.