QBSQ: A Quad-Tree Based Algorithm for Skyline Query

Skyline has been proposed as an important operation for multi-criteria decision making, data mining and visualization, and user preference queries. In this paper, we systematically explore a quad-tree based approach for computing skyline points which contributes to a better performance than traditional ones for skyline queries. Based on this approach, we present a novel algorithm QBSQ for finding the set of global skyline points. QBSQ partitions data points dynamically by means of the configuration characters of quad-tree, and then deletes points dominated by other(s) while constructing the tree. The amount of work for domination checking is minimized and the efficiency of our algorithm is thus improved. Extensive experiments prove the efficiency and the scalability of QBSQ algorithm.