Reverse Skyline Search in Uncertain Databases

Reverse skyline queries over uncertain databases have many important applications such as sensor data monitoring and business planning. Due to the wide existence of uncertainty in many real-world data, answering reverse skyline queries accurately and efficiently over uncertain data has become increasingly important. In this article, we formalize the probabilistic reverse skyline query over uncertain data, in both monochromatic and bichromatic cases, and propose effective pruning methods, namely spatial pruning and probabilistic pruning, to reduce the search space of the reverse skyline query processing. Moreover, efficient query procedures have been presented seamlessly integrating the proposed pruning methods. Furthermore, a novel query type, namely Probabilistic Reverse Furthest Skyline (PRFS) query, is proposed and tackled under "the larger, the better" dominance semantics of skyline. Variants of probabilistic reverse skyline have been proposed and tackled, including those that return objects with top-k highest probabilities and that retrieve top-k reverse skylines. Extensive experiments demonstrated the efficiency and effectiveness of our approaches with various experimental settings