Flexible user-centric service selection algorithm for Internet of Things services

Abstract

In Internet of Things (IoT), the similar functional services are evolving in different quality of services (QoS) due to the widespread deployment of spatially distributed things on dynamic networks through the web. Therefore, the user involvement in service selection systems becomes a vital role to enhance the system performance by taking into account of subjective factors. This paper proposes a flexible QoS-based service selection algorithm (FQSA) which mainly targets to the users to be able to give their subjective preferences in an easy and friendly manner. The FQSA algorithm selects the services based on subjective information provided by the service consumers and objective information supported by the service providers. It adopts artificial neural network backpropagation algorithm to find the objective factors and applies similarity aggregation method to evaluate the creditability of the user subjective factors which have already been evaluated by ontological reasoning with the help of proposed QoS ontology. The comparative study and experimental results show that our FQSA algorithm is superior to other service selection approaches.