Given a spatial data set placed on an n x n grid, our goal is to find the rectangular regions within which subsets of the data set exhibit anomalous behavior. We develop algorithms that, given any use
