Efficient Distributed Skylining for Web Information Systems

Abstract

Abstract. Though skyline queries already have claimed their place in retrieval over central databases, their application in Web information systems up to now was impossible due to the distributed aspect of retrieval over Web sources. But due to the amount, variety and volatile nature of information accessible over the Internet extended query capabilities are crucial. We show how to efficiently perform distributed skyline queries and thus essentially extend the expressiveness of querying today's Web information systems. Together with our innovative retrieval algorithm we also present useful heuristics to further speed up the retrieval in most practical cases paving the road towards meeting even the realtime challenges of on-line information services. We discuss performance evaluations and point to open problems in the concept and application of skylining in modern information systems. For the curse of dimensionality, an intrinsic problem in skyline queries, we propose a novel sampling scheme that allows to get an early impression of the skyline for subsequent query refinement.

Multi-objective Query Processing for Database Systems

“…We will also use two variables a i and b i with each list K i , such that all objects in K i having larger scores than a i with respect to f i have already been output or discarded as dominated and b i is the current value of f i using the lowest scores seen so far as input. That means in every K i we are done with objects having higher or equal values of f i than a i , and we have to consider objects for output having values between a i and b i and cannot yet output all objects with values up to b i (for a detailed discussion of a i and b i see [1]). Furthermore we will use two sets R and X containing the output optimal result objects and the discarded dominated objects respectively.…”

“…Examples for this are e.g. the middleware top k retrieval in [11] and the central index top k retrieval in [8] or the distributed skylining in [1] and the central index skylining in [23]. However, in all these cases basic algorithms are similar.…”

“…[5], [24], [14], [23]. Also in this case the algorithms differ in a variety of heuristics and techniques used, but again we will present the basic optimal algorithm only using the abstract access methods over sorted lists for each aspect as given in [1].…”

“…■ We will now state that both basic algorithms for top k retrieval and skylining are just special cases of our multiobjective retrieval algorithm. Please note that the respective algorithms were proven to be optimal in [10] and [1].…”

Finding Key Persons on Social Media by Using MapReduce Skyline

“…If the datasets are too large and stored in a distributed environment, it may impossible to handle them in a centralized fashion [12]. Balke et al first excavated skyline in a distributed environment by partitioning the whole dataset vertically [2]. Vlachou et al introduced the concept of extended skyline set, which contains all data elements that are necessary to answer a skyline query in any arbitrary subspace [24].…”