Sampling-based approximate skyline calculation on big data

Nowadays, big data is coming to the force in a lot of applications. Processing skyline query on more than linear time by far too expensive and often even may be slow. It obviously not possible compute an exact solution sublinear time, since itself have size. Fortunately, many situations, fast approximate useful slower solution. This paper proposes two sampling-based algorithms for processing queries. The first algorithm obtains fixed size sample computes it. error only relatively small most cases, but also almost unaffected input second returns [Formula: see text]-approximation efficiently. running has nothing do with practical, achieving goal sublinearity data. Experiments verify analysis algorithm, show that much faster existing algorithms.