SkyCover: Finding Range-Constrained Approximate Skylines with Bounded Quality Guarantees

Skyline queries retrieve promising data objects that are not dominated in all the attributes of interest. However, in many cases, a user may not be interested in a skyline set computed over the entire dataset, but rather over a specified range of values for each attribute. For example, a user may look for hotels only within a specified budget and/or in a particular area in the city. This leads to constrained skylines. Even after constraining the query ranges, the size of the skyline set can be impractically large, thereby necessitating the need for approximate or representative skylines. Thus, in this paper, we introduce the problem of finding range-constrained approximate skylines. We design a grid-based framework, called SkyCover, for computing such skylines. Given an approximation error parameter > 0, the SkyCover framework guarantees that every skyline is “covered” by at least one representative object that is not worse by more than a factor of (1 + ) in all the dimensions. This is achieved by employing a non-uniform grid partitioning on the data space. We also propose two new metrics based on the covering factor to assess the quality of an approximate skyline set. Experimental evaluation reveals that SkyCover outperforms the competing methods in both quality and running time.