A linguistic quantifier-based approach for skyline refinement
Katia Abbaci, Allel Hadjali, Ludovic Liétard, Daniel Rocacher · 2013
Skyline queries are a popular and powerful paradigm for extracting interesting objects from a d-dimensional dataset. They rely on Pareto dominance principle to identify the skyline objects, i.e., the set of incomparable objects which are not dominated by any other object from the dataset. Two main problems may be faced when using skyline queries: (i) a small number of returned objects which could be insufficient for the users' needs; and (ii) a huge number of skyline objects which is less informative for the users. In this paper, we tackle the last problem and propose an approach to deal with it. The idea consists in refining the skyline in order to discriminate its elements and select the best ones. A new definition of dominance relationship based on the fuzzy quantifier “almost all” is then introduced.