A Privacy-sensitive Service Selection Method Based on Artificial Fish Swarm Algorithm in the Internet of Things

Compared with the traditional Internet services, the services under the Internet of Things (IoT) expand the binary field of “user and information space” to the ternary field of “user, information space, and physical space”. How to aggregate all kinds of information, contents and applications, and how to filter services according to users’ demands, especially the privacy requirements, have become a key issue in IoT applications. Most of the existing service selection algorithms do not consider privacy factors and just adopt simple methods such as Heuristic Search Algorithm (HSA) and Genetic Algorithm (GA etc). HSA relies too much on heuristic strategies which may cause unstable performance, while GA cannot well meet the needs of service selection with multi-path characteristics due to the one-dimensional chromosome coding. To overcome the disadvantages above, this paper propose a privacy-sensitive service selection algorithm based on the Artificial Fish Swarm Algorithm (ASFA). It aims to choose the service with the best Quality of Experience (QoE) which includes privacy preferences as one of its primary factors so as to reduce the risk of privacy exposure and to pick up the service that satisfies all the requirements of users. Specifically, QoE model with privacy preferences is established and relevant constraints as well as quantitative methods are given firstly. Secondly, the proposed algorithm is constructed to select specific services based on the above model. Finally, the proposed method is verified through simulations. The results show that, compared with the GA-based algorithm, the proposed algorithm has improved both the precision rate and recall rate by more than 10% on average, which means it can solve the privacy-sensitive service selection problems in IoT with a feasible and effective way.