A novel approach for selecting the top skyline under users' references

Skyline query is useful in many multi-criteria decision making applications. However, skyline queries on large data sets usually produce too many results to offer interesting insights. In this paper, we propose a new query called referential skyline, which refines the skyline according to a few reference points provided by the user. Given a reference, a referential skyline (RS) query retrieves the skyline points which dominate it. When given multiple references, the skyline points are ranked by the number of references that each one dominates. We present an R-tree based scheme to process the referential skyline query. We also propose the problem to retrieve the Top-δ Referential Skyline (TRS) under multiple references. Our comprehensive experiments indicate the effectiveness and efficiency of our scheme when processing referential skyline and top-δ referential skyline queries.