hey folks, is there an efficient index or aggregation function for Top-N queries? Approximations would be fine. Thanks!