arXiv 2207.06819
Anomal-E: A Self-Supervised Network Intrusion Detection System based on Graph Neural Networks
By Evan Caville, Wai Weng Lo, et al.
Published 2022-07-14
Wiki summary
Explore the paper's summary, context, and related research on Papiers.
This paper investigates Graph Neural Networks (GNNs) application for self-supervised network intrusion and anomaly detection. GNNs are a deep learning approach for graph-based data that incorporate graph structures into learning to generalise graph representations and output embeddings. As network flows are naturally graph-based, GNNs are a suitable fit for analysing and learning network behaviour. The majority of c…