Graph Convolutional Networks Improve Cloud Phishing Detection Across Device Interactions

A computational evaluation demonstrates that graph convolutional networks can detect device-aware phishing attacks targeting cloud-based systems with high precision. Researchers developed a detection approach structured around graph-based learning to map complex relationships linking users, connected devices, and suspect activities. Cloud platforms remain vulnerable to evolving fraudulent techniques that bypass conventional signature-based security filters, creating a need for network architectures capable of recognizing subtle behavioral interactions across distributed infrastructure.

The study implemented graph convolutional networks to model device interactions and identify malicious indicators that traditional defense systems miss. In experimental testing, the proposed model attained an accuracy of 99.94 percent along with a precision rate of 99.91 percent. The evaluation also recorded a sensitivity of 99.94 percent, a specificity of 99.94 percent, and an overall F-measure reaching 99.93 percent across the monitored cloud environment configurations.

Comparative evaluations against established baseline algorithms confirmed the performance advantages of the graph-based framework. The proposed model outperformed several benchmark architectures, specifically MLSELM, ECODL-IDSC, and a hybrid recurrent neural network combining gated recurrent units, across standardized metrics covering accuracy, precision, recall, and F1-score determinations.

Despite the high classification rates, the researchers noted operational boundaries that highlight targets for technical refinement. The system recorded a false positive rate of 0.06 percent alongside a corresponding false negative rate of 0.06 percent during testing. The investigators reported that addressing these residual classification errors will provide a foundation for advancing cybersecurity detection models in cloud environments.

Subject of Research: Graph convolutional networks for cloud-based phishing detection

Article Title: Graph convolutional networks (GCNs) for detecting device-aware phishing attacks in cloud environments

Article References: Mubarakali, A., Alqahtani, A. S., Elshafie, H., Changalasetty, S. B., & Al Hanif, A. (2026). Graph convolutional networks (GCNs) for detecting device-aware phishing attacks in cloud environments. Cluster Computing, 29(16), Article 938. https://doi.org/10.1007/s10586-026-06749-8

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06749-8

Keywords: graph convolutional networks, phishing detection, cloud computing, cybersecurity, device-aware security, machine learning, network security, signature-based detection, MLSELM, ECODL-IDSC, RNN-GRU, false positive rate, false negative rate, F-measure, graph-based learning, cloud platforms