Design And Analysis Of The Artificial Bee Colony Algorithms For Domain-Oriented Qos-Aware Service Composition
Haifang Wang,Xiaofei Xu,Zhongjie Wang,Zhizhong Liu
DOI: https://doi.org/10.1504/ijstm.2016.10000550
2016-01-01
International Journal of Services Technology and Management
Abstract:Recently, various evolutionary algorithms have been successfully applied to acquire approximately optimal solutions for QoS-aware service composition problems. Especially, artificial bee colony algorithm (ABC) stands out due to its advantages of few parameters, strong robustness and search capability. In addition, for these algorithms, domain features are widely utilised as heuristic. However, how to combine these two points to achieve more optimal solutions is becoming a challenge. To address this critical challenge, this paper summarises three domain features (priori, similarity and correlation) and proposes the artificial bee colony algorithms for domain-oriented service composition named S-ABCsc paradigm. Besides, the framework of the paradigm is presented, and its seven configurable points are identified. Moreover, two types of algorithms are derived from the paradigm. To apply the paradigm, a support tool is realised, which helps users obtain a concrete algorithm. Furthermore, several comparison experiments are conducted, which have proved the effectiveness of the paradigm.