Research Article
“Failure” Service Pattern Mining for Exploratory Service Composition
@INPROCEEDINGS{10.1007/978-3-030-92635-9_3, author={Yunjing Yuan and Jing Wang and Yanbo Han and Qianwen Li and Gaojian Chen and Boyang Jiao}, title={“Failure” Service Pattern Mining for Exploratory Service Composition}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 17th EAI International Conference, CollaborateCom 2021, Virtual Event, October 16-18, 2021, Proceedings, Part I}, proceedings_a={COLLABORATECOM}, year={2022}, month={1}, keywords={Exploratory service composition Service pattern mining gSpan}, doi={10.1007/978-3-030-92635-9_3} }- Yunjing Yuan
Jing Wang
Yanbo Han
Qianwen Li
Gaojian Chen
Boyang Jiao
Year: 2022
“Failure” Service Pattern Mining for Exploratory Service Composition
COLLABORATECOM
Springer
DOI: 10.1007/978-3-030-92635-9_3
Abstract
To adapt to uncertain and dynamic requirements, exploratory service composition enables business users to construct service composition processes in a trial-and-error manner. A large number of service composition processes are generated, which can be learned to improve the reusability of the service composition processes. By mining these service composition processes and abstracting the mining results to service patterns, the efficiency of service composition can be effectively improved. At present, there has been some research on the methods of service pattern mining. Most of the work focuses on successful service composition processes, but the ones that fail are also valuable. For example, by using the mining results of failure service composition processes, the accuracy of service recommendations can be improved. To solve this problem, this paper proposes a “failure” service pattern mining algorithm (FSPMA) for exploratory service composition, which extends the gSpan algorithm, and can mine “failure” service patterns from service composition processes for further reuse. Meanwhile, the exploratory service composition model and the service pattern model are explained for the FSPMA. The prototype implementation of the exploratory service composition environment is introduced, which integrates the FSPMA. The experimental evaluation is explained to verify the algorithm, and the result shows that the efficiency of the FSPMA has a significant improvement in mining “failure” service patterns compared with the gSpan algorithm and the TKG algorithm. Finally, the application of “failure” service patterns in service recommendations is given.
