Hybrid immune-inspired method for selecting the optimal or a near-optimal service composition

The increasing interest in developing optimization techniques that provide the optimal or a near-optimal solution of a problem in an efficient way has determined researchers to turn their attention towards biology. It has been noticed that biology offers many clues regarding the design of such optimization techniques, since biological systems exhibit self-optimization and self-organization capabilities in a decentralized way without the existence of a central coordinator. In this context we propose a bio-inspired hybrid method that selects the optimal or a near-optimal solution in semantic Web service composition. The proposed method combines principles from immune-inspired, evolutionary, and neural computing to optimize the selection process in terms of execution time and explored search space. We model the search space as an Enhanced Planning Graph structure which encodes all the possible composition solutions for a given user request. To establish whether a solution is optimal, the QoS attributes of the services involved in the composition as well as the semantic similarity between them are considered as evaluation criteria. For the evaluation of the proposed selection method we have implemented an experimental prototype and carried out experiments on a set of scenarios from the trip planning domain.