Optimizing cloud service cryptography via fuzzy graph theory neural networks: A data model perspective

Abstract

A new framework integrates graph neural networks (GNNs) and fuzzy graph theory (FGT) to improve cloud service selection cryptography. Nodes represent cloud services, and edges indicate trust relations with fuzzy weights in the FGT cloud service ecosystem. GNN captures complex ecosystem dependencies and uncertainty. Imprecise trust measurements are handled using fuzzy logic. Service selection is optimized via GNN-based message forwarding while complying with cryptographic limitations. The suggested method supports real-time cloud systems by accounting for dynamic trust relations. Fuzzy logic handles inaccurate trust measures, and GNN-based message forwarding optimizes service selection and security analysis under cryptographic limitations in the proposed approach. This study tests a synthetic cloud service dataset. The framework proves stability in dynamic trust relationships and large-scale cloud scalability analysis in AWS (Amazon Web Services), Microsoft Azure, Google Cloud Platform (GCP), Oracle Cloud Infrastructure (OCI), and IBM (International Business Machines). Experimental findings show that the GCP outperforms previous techniques by 97% in efficiency and 10% in security measures, with 95% service selection accuracy and a 2.8% false positive rate. A synthetic cloud service dataset validates the framework’s resilience in dynamic trust relationships and scalability for large cloud settings.