Sampling Based Approximate Skyline Calculation on Big Data

TL;DR

This paper introduces two sampling-based approximate algorithms for skyline queries on big data, significantly reducing computation time while maintaining acceptable accuracy, thus enabling scalable skyline analysis.

Contribution

It presents novel sampling algorithms for approximate skyline computation that are faster and scalable compared to existing exact methods.

Findings

The first algorithm has small error, nearly independent of data size.

The second algorithm provides an $(psilon,elta)$-approximation with constant sample size.

The second algorithm is much faster than existing skyline algorithms.

Abstract

The existing algorithms for processing skyline queries cannot adapt to big data. This paper proposes two approximate skyline algorithms based on sampling. The first algorithm obtains a fixed size sample and computes the approximate skyline on the sample. The error of the first algorithm is relatively small in most cases, and is almost independent of the input relation size. The second algorithm returns an $(\epsilon,\delta)$-approximation for the exact skyline. The size of sample required by the second algorithm can be regarded as a constant relative to the input relation size, so is the running time. Experiments verify the error analysis of the first algorithm and show that the second algorithm is much faster than the existing skyline algorithms.