CAGN-GAT Fusion: A Hybrid Contrastive Attentive Graph Neural Network for Network Intrusion Detection

TL;DR

This study tackles network intrusion detection in resource-constrained environments by proposing CAGN-GAT Fusion, a hybrid model that combines Contrastive Attentive Graph Networks (CAGN) with Graph Attention Networks (GAT) in a dual-stream architecture. It introduces adaptive graph construction and graph augmentation to shape graph representations and evaluate performance across four datasets with a fixed $5000$ samples, benchmarking against 15 baselines. The results show that CAGN-GAT Fusion delivers competitive accuracy, recall, and F1, with strong efficiency, while revealing nuanced effects of graph density and augmentation on detection performance. The work sheds light on when contrastive learning benefits graph-based IDS and points to future improvements through GraphSAGE integration and multiview/dynamic graph constructions for enhanced robustness and scalability.

Abstract

Cybersecurity threats are growing, making network intrusion detection essential. Traditional machine learning models remain effective in resource-limited environments due to their efficiency, requiring fewer parameters and less computational time. However, handling short and highly imbalanced datasets remains challenging. In this study, we propose the fusion of a Contrastive Attentive Graph Network and Graph Attention Network (CAGN-GAT Fusion) and benchmark it against 15 other models, including both Graph Neural Networks (GNNs) and traditional ML models. Our evaluation is conducted on four benchmark datasets (KDD-CUP-1999, NSL-KDD, UNSW-NB15, and CICIDS2017) using a short and proportionally imbalanced dataset with a constant size of 5000 samples to ensure fairness in comparison. Results show that CAGN-GAT Fusion demonstrates stable and competitive accuracy, recall, and F1-score, even though it does not achieve the highest performance in every dataset. Our analysis also highlights the impact of adaptive graph construction techniques, including small changes in connections (edge perturbation) and selective hiding of features (feature masking), improving detection performance. The findings confirm that GNNs, particularly CAGN-GAT Fusion, are robust and computationally efficient, making them well-suited for resource-constrained environments. Future work will explore GraphSAGE layers and multiview graph construction techniques to further enhance adaptability and detection accuracy.