|
|
 |
| |
|
|
|
A Novel Enhancement of Bird Swarm Algorithm for Efficient Web Service Composition |
|
|
|
PP: 215-226 |
|
|
doi:10.18576/amis/200114
|
|
|
|
Author(s) |
|
|
|
Fadl Dahan,
|
|
|
|
Abstract |
|
|
| We propose an optimized version of Bird Swarm Algorithm (BSA), termed Improved Bird Swarm Algorithm (IBSA) to solve Web Service Composition (WSC) problem. WSC plays a key role in fulfilling complex user requests, especially Quality of Services (QoS) categories. By its nature, WSC is a multi-objective optimization problem and is NP-hard in computing. To deal with these challenging problems, optimization algorithms inspired by swarm intelligence have been investigated. Our proposed algorithm improves the BSA performance using three mechanisms Firstly, the Sine Chaos Theory has been proposed to initiate the population of BSA and to amplify the diversity within the initialization phase. Secondly, a Le ́vy flight mechanism has been introduced to improve the dual aspects of exploitation and exploration within BSA by preserving the diversity of the avian agents. Furthermore, a neighborhood search mechanism has been employed to mitigate the trade-off between exploration and exploitation within searching methodologies. The efficacy of the proposed approach is assessed utilizing both real and synthetic web service composition datasets (encompassing a cumulative total of 72,000 web services) and compared with standard BSA and three state-of-the-art algorithms, with findings demonstrating its superiority relative to other contemporary methodologies in terms of composition quality and computational efficiency. Consequently, the suggested methodology offers a pragmatic and efficient resolution to the web service composition dilemma, signifying a substantial progression in the domain of service-oriented computing. |
|
|
|
|
|
|
|
