UCOS: Enhanced Online Skyline Computation by User Clustering

Author

Kehan Chen ; Lichuan Ji ; Kunyang Jia ; Jian Wu

Author_Institution

Coll. of Comput. Sci. & Technol., Zhejiang Univ., Hangzhou, China

fYear

2013

fDate

June 28 2013-July 3 2013

Firstpage

763

Lastpage

764

Abstract

In this paper, we propose a skyline computation system UCOS (User Clustering based Online Skyline), which divides the computation into offline and online stages. Based on the truth that QoS similarity implies the skyline similarity, the offline stage of UCOS system dose user clustering according to the historical user-service QoS records by given distance metrics. Then, we compute the representative skyline for each cluster standing for the general characters of the users´ skylines. Benefit from those offline results, the online stage is able to give a rapid prediction for online skyline request and achieves good online computation performance by doing refinement on the predicted results.

Keywords

Web services; pattern clustering; quality of service; QoS similarity; UCOS skyline computation system; distance metrics; historical user-service QoS records; offline computation stage; online computation stage; online skyline request; quality of service; user clustering based online skyline; Clustering algorithms; Computer architecture; Conferences; Educational institutions; Measurement; Prediction algorithms; Quality of service; Clustering; QoS; Service Selection; Skyline;

fLanguage

English

Publisher

ieee

Conference_Titel

Services Computing (SCC), 2013 IEEE International Conference on

Conference_Location

Santa Clara, CA

Print_ISBN

978-0-7695-5026-8

Type

conf

DOI

10.1109/SCC.2013.14

Filename

6649774