Steered by end users, data services-based Internet data instant integration applies data services as the unified abstraction of data sources on the Internet. End users can integrate data from multiple sources through composing services, according to their instant and personalized requirements. However, it is still challenging for users to choose and compose services manually, which may lead to poor precision and relatively low efficiency. Towards this issue, we first introduce a data service hyperlink model to fully represent uni-directional correlations between input and output parameters of data services. In order to assist users to compose data services in a rapid and accurate way, this paper proposes a method to reuse correlations of two services, in which the composition plan is generated automatically according to the service hyperlink between them. In this way, the difficulty of composing services is greatly reduced. Experimental results show that the proposed method improves the efficiency and accuracy of the Internet data instant integration to a great extent.
|Number of pages||14|
|Journal||International Journal of Grid and Distributed Computing|
|Publication status||Published - 1 Jan 2017|
- Automatic service composition plan generation
- Data correlation
- Data service
- Instant data integration