Paper Info
Title
Temporal Pattern Based QoS Prediction.
Abstract
Quality-of-Service QoS is critical for selecting the optimal Web service from a set of functionally equivalent service candidates. Since QoS performance of Web services are unfixed and highly related to the service status and network environments which are variable against time, it is critical to obtain the missing QoS values of candidate services at given time intervals. In this paper, we propose a temporal pattern based QoS prediction approach to address this challenge. Clustering approach is utilized to find the temporal patterns based on services QoS curves over time series, and polynomial fitting function is employed to predict the missing QoS values at given time intervals. Furthermore, a data smoothing process is employed to improve prediction accuracy. Comprehensive experiments based on a real world QoS dataset demonstrate the effectiveness of the proposed prediction approach.
Year
DOI
Venue
2016
10.1007/978-3-319-48743-4_18
WISE
Keywords
Field
DocType
Service Computing, QoS prediction, Temporal pattern
Services computing,Data mining,Polynomial,Computer science,Quality of service,Smoothing,Web service,Cluster analysis
Conference
Volume
ISSN
Citations 
10042
0302-9743
0
PageRank 
References 
Authors
0.34
18
6
Name
Order
Citations
PageRank
Liang Chen125828.02
Haochao Ying27310.03
Qibo Qiu300.34
Jian Wu493395.62
Hai Dong543941.61
Athman Bouguettaya62579267.41