Supporting personalized top-k skyline queries using partial compressed skycube

摘要: As near-infinite amount of data are becoming accessible on the Web, it is getting more and important to support intelligent query mechanisms, help each user identify ideal results manageable size. such mechanism, skyline queries have gained a lot attention lately for its intuitive formulation. This intuitiveness, however, has side-effect generating too many results, especially high-dimensional data, satisfy wide range user's needs. Our goal personalized as identifying "truly interesting" objects based user-specific preference retrieval size k. While this problem been studied previously, proposed solution identifies top-k by navigating "skycube", which incurs exponential storage overhead dimensionality excessive one-time computational skycube construction. In contrast, we develop novel techniques significantly reduce both computation overhead. extensive evaluation validate framework real-life synthetic data.