Skyline Queries Computation on Crowdsourced- Enabled Incomplete Database

摘要: Data incompleteness becomes a frequent phenomenon in large number of contemporary database applications such as web autonomous databases, big data, and crowd-sourced databases. Processing skyline queries over incomplete databases impose challenges that negatively influence processing the queries. Most importantly, skylines derived from are also which some values missing. Retrieving with missing is undesirable, particularly, for recommendation decision-making systems. Furthermore, running on data raises issues losing transitivity property technique cyclic dominance between tuples. The issue estimating has been discussed examined literature. recently, several studies have suggested exploiting order to estimate by generating plausible using crowd. Crowd-sourced proved be powerful solution perform user-given tasks integrating human intelligence experience process tasks. However, task incurs additional monetary cost increases time latency. Also, it not always possible produce satisfactory result meets user’s preferences. This paper proposes an approach first available utilizes implicit relationships attributes impute skylines. aims at reducing estimated crowd when local estimation inappropriate. Intensive experiments both synthetic real datasets accomplished. experimental results proven proposed enabled scalable outperforms other existing approaches.