Warning: ini_set(): A session is active. You cannot change the session module's ini settings at this time in /home/medicalscie/public_html/thesesjournal.com/lib/pkp/classes/session/SessionManager.inc.php on line 69

Warning: Cannot modify header information - headers already sent by (output started at /home/medicalscie/public_html/thesesjournal.com/lib/pkp/classes/session/SessionManager.inc.php:69) in /home/medicalscie/public_html/thesesjournal.com/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.inc.php on line 478

Warning: Cannot modify header information - headers already sent by (output started at /home/medicalscie/public_html/thesesjournal.com/lib/pkp/classes/session/SessionManager.inc.php:69) in /home/medicalscie/public_html/thesesjournal.com/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.inc.php on line 479
TY - JOUR AU - M Tahseen Alam, AU - Muhammad Waseem, AU - Hafiza Iqra Iftikhar, PY - 2026/03/10 Y2 - 2026/10/08 TI - NEUROFUSION-X: A HYBRID TRANSFORMER–GNN DEEP LEARNING MODEL FOR PROACTIVE CYBER-ATTACK PREDICTION IN IOT NETWORKS JF - Spectrum of Engineering Sciences JA - SES VL - 4 IS - 3 SE - Articles DO - UR - https://www.thesesjournal.com.medicalsciencereview.com/index.php/1/article/view/2170 SP - 211-232 AB - <p><em>The rapid expansion of Internet of Things (IoT) networks has introduced significant security challenges due to the heterogeneous, large-scale, and highly dynamic nature of IoT traffic. Contemporary intrusion detection systems (IDS) predominantly rely on traditional machine learning or single-architecture deep learning models, which are largely reactive and insufficient for capturing the complex spatio-temporal characteristics of modern cyber-attacks. In particular, existing approaches often fail to jointly model long-range temporal dependencies and network-level communication topology, limiting their effectiveness in proactive threat mitigation.</em></p><p><em>This paper proposes <strong>NeuroFusion-X</strong>, a hybrid deep learning framework that integrates Transformer-based temporal modeling with Graph Neural Network (GNN)-based relational learning for <strong>proactive multi-class cyber-attack prediction</strong> in IoT networks. The proposed architecture leverages self-attention mechanisms to learn evolving attack patterns across time while simultaneously capturing inter-device communication dependencies through graph-based message passing. A fusion module unifies temporal and topological representations to forecast future attack categories before full attack execution, shifting intrusion detection from reactive classification to predictive cyber defense.</em></p><p><em>Extensive experiments are conducted on large-scale IoT security datasets, including <strong>BoT-IoT</strong> and <strong>CICIoT2023</strong>, encompassing diverse attack types and protocol behaviors. Experimental results demonstrate that NeuroFusion-X consistently outperforms traditional machine learning models, Transformer-only, and GNN-only baselines in terms of macro-F1 score, ROC-AUC, and early prediction accuracy, particularly under severe class imbalance. The findings confirm that spatio-temporal fusion significantly enhances attack separability, prediction stability, and proactive detection capability. Overall, NeuroFusion-X provides a scalable, extensible, and future-ready framework for intelligent cyber-defense in next-generation IoT infrastructures.</em></p> ER -