Predicting service collaboration for users based on data variation patterns Online publication date: Mon, 12-Oct-2020
by Jiaqiu Wang; Zhongjie Wang
International Journal of Information Technology and Management (IJITM), Vol. 19, No. 4, 2020
Abstract: Service collaboration allows the realisation of more complicated business logic by using existing services. Nowadays, users use lots of services in their daily work. For example, developers use a large number of services (e.g., Stack Overflow, Github, Blogger, etc.) to develop programs. Services are used continuously. Since most of the users' data is distributed in these different service providers, these data are separated from each other although they are correlated. If we coordinate different services based on these correlation data, we can provide users with seamless and effective support. This is very significant because it greatly increases users' productivity. However, due to the segregation of data, it is difficult to coordinate different services based on data correlation. To deal with this challenge, we propose a novel deep recurrent neural network (runs in a centralised service) to predict future services collaboration and their generated data. The network captures the correlation between different data and discovers patterns of data variation by using multiple hidden layers, which are beneficial to services collaboration prediction. Extensive experiments are conducted on the real world dataset. Experimental results show that our model significantly outperforms a few competitive baseline methods.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Information Technology and Management (IJITM):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com