Title: A QoS prediction approach based on fusion of network representation learning and dynamic collaborative filtering for cloud service

Authors: LiQiong Chen; GuoQing Fan; Kun Guo

Addresses: Department of Computer Science and Information Engineering, Shanghai Institute of Technology, No. 100, Hai Quan Road, Feng Xian District, 201418 Shanghai, China ' Department of Computer Science and Information Engineering, Shanghai Institute of Technology, No. 100, Hai Quan Road, Feng Xian District, 201418 Shanghai, China ' Department of Computer Science and Information Engineering, Shanghai Institute of Technology, No. 100, Hai Quan Road, Feng Xian District, 201418 Shanghai, China

Abstract: In recent years, the number of cloud services has increased dramatically with the development of internet technology. The recommendation of cloud service based on quality of service (QoS) has received more and more attention. The traditional matrix decomposition model has cold start and matrix sparse problems, which cannot fully describe the relationship between users and services, thus affecting the effect of QoS value prediction. In order to accurately predict the QoS value of missing cloud services, this paper proposes a prediction method of fusion network representation learning and dynamic collaborative filtering. The method can better mine the high-dimensional relationship vector of the user-service-attribute ternary network feature through network representation learning. The high-dimensional relation vector learned by the network representation is integrated into the dynamic collaborative filtering algorithm to solve the problem of sparse QoS matrix data in the dynamic collaborative filtering method. The effectiveness of the proposed method is verified by experimental simulation.

Keywords: cloud service; QoS prediction; collaborative filtering; network representation learning; personalised recommendation.

DOI: 10.1504/IJIITC.2021.115708

International Journal of Intelligent Internet of Things Computing, 2021 Vol.1 No.3, pp.184 - 199

Received: 22 Nov 2019
Accepted: 30 Jan 2020
Published online: 17 Jun 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article