Learning automata-based adaptive web services composition
- 1 June 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 23270586,p. 792-795
Abstract
Service-oriented computing is a widely adopted paradigm in real applications. Considering the continuous evolution of services, adaptive service composition has always been a major concern. It is a big challenge to adjust the composition to be optimal in real-time. In this paper, a learning automata-based approach is proposed to attack this problem. It consists of two important components: random environment and a learning automaton. The former can be mapped to the service's execution environment. The latter is responsible for the adaptation achievement using reward and penalty functions, while we take the service composition structures into account to compute the usefulness value of all services. At last, simulation study has shown that our approach is efficient to find the optimal (sub-optimal) composition.Keywords
This publication has 16 references indexed in Scilit:
- A learning automata-based fault-tolerant routing algorithm for mobile ad hoc networksThe Journal of Supercomputing, 2011
- A Penalty-Based Approach for QoS Dissatisfaction Using Fuzzy Rules2011
- A survey on self-healing systems: approaches and systemsComputing, 2010
- FACTS: A Framework for Fault-Tolerant Composition of Transactional Web ServicesIEEE Transactions on Services Computing, 2009
- Personalized Web Service Ranking via User Group Combining Association Rule2009
- A QoS-aware fault tolerant middleware for dependable service composition2009
- Discovering the best web service2007
- Adaptive Service Composition in Flexible ProcessesIEEE Transactions on Software Engineering, 2007
- A Fault-Tolerant Routing Algorithm for Mobile Ad Hoc Networks Using a Stochastic Learning-Based Weak Estimation Procedure2006
- Learning Automata - A SurveyIEEE Transactions on Systems, Man, and Cybernetics, 1974