Explanations for Skyline Query Results

摘要: Skyline queries are a well-studied problem for multidimensional data, wherein points returned to the user i↵ no other point is preferable across all attributes. This leaves only most likely appeal an arbitrary user. However, some dominated may still be interesting, and skyline o↵ers little support helping understand why interesting omitted from results. In this paper, we introduce Sky-not query .G iven p, dataset S, constraints with bounding corners qL qU, returns alternative q 0 closest which p in skyline. equips understanding of not just that was dominated, but also how severely. He can then assess himself whether competitive. We first propose theoretical results show drastically reduce input processed by query, independent any algorithm. o↵er skyline-like ecient recursive algorithm solving queries, evaluate extensive experimental evaluation.