The main contribution of this work is the development of
the TOP-k COrrelated-Pairs (TOP-COP) query algorithm.
We show that TOP-COP finds the top-k correlated pairs in a
computation time which can be orders of magnitude faster
than the brute-force alternative. The algorithm exploits a 2-
D monotone property of the upper bound of Pearson’s correlation
coefficient. By interpreting this property geometrically,
we obtain an algorithm that uses a diagonal traversal
method, combined with a refine-and-filter strategy, to efficiently
find the top-k pairs.