Representative skylines using threshold-based preference distributions

(Approximate) Uncertain Skylines

“…In higher dimensions, providing approximate levels of uncertain skylines with guarantees remains a challenge (as with similar problems on precise data [16]). A similar argument for the importance of concise skylines is made regarding work on k-dominant skylines [6] and other formulations [24,12]; the difference is that our work provides approximation guarantees in the value of the attributes and operates on data that has defined uncertainty.…”

Efficient algorithms for finding the most desirable skyline objects

“…More recently, Lee and Hwang [24] develop an efficient greedy algorithm for the k representative skyline using the skytree. As proved in [11,30,37,41], the problem of representative skyline retrieval is NP-hard for the dimensionality at least 3. Compared with the representative skyline, the MDSO query employs a new ranking criterion defined in Definition 6, and it is not NP-hard in high dimensional spaces.…”

“…Vlachou et al [41] present a novel framework for discovering the representative skyline over distributed data sources, which incorporates the above two metrics, i.e., dominance-based representative [30] and distance-based representative [37]. Das et al [11] define a new k representative skyline object by ensuring that the probability which a random user would click on one of them is maximized. More recently, Lee and Hwang [24] develop an efficient greedy algorithm for the k representative skyline using the skytree.…”

Computing Skyline from Evidential Data

“…In the years that followed the emergence of the concept of skyline queries, computing the skyline was the major concern, most of the works were about designing efficient evaluation algorithms under different conditions and in different contexts, see for instance [9] [10] [29] [13]. In the last decade, skyline computation over uncertain data has also attracted the interest of many researchers.…”