Due the increasing number of published Semantic Web services (SWs) rendered the distributed discovery within repositories a critical issue and a major problem that can reduce the capability and functionality of SWs in terms of efficiency and scalability. Peer-to-Peer (P2P) computing is considered as the most dominant technology to discover new distributed and heterogeneous collaborative applications for SWs. In this paper, we propose an efficient approach for improving the performance and effectiveness of automatic and cooperative discovery of large-scale distributed systems in the unstructured P2P networks. The approach exploits a scalable epidemic algorithm that uses different sources of network knowledge, such as exponential distribution, to fulfill the users requirements in order to ensure high recall, further reduce the number of messages exchanged and reduce the execution time for discovering SWs in the unstructured P2P network. In order to improve the applicability of the scalable epidemic algorithm for discovering SWs, we propose the semantic matching of OWL-S process model which improves the recall while keeping an acceptable matching quality level. The experimental results show that our efficient approach is able dynamically to adapt to network changes and preserve high levels of recall.
Boukhadra, A., Benatchba, K., & Balla, A. (2015). Similarity flooding for efficient distributed discovery of OWL-S process model in P2P networks. In Procedia Computer Science (Vol. 56, pp. 317–324). Elsevier B.V. https://doi.org/10.1016/j.procs.2015.07.214