One-Class Intrusion Detection with Dynamic Graphs
This work addresses network security challenges for digital systems, but it appears incremental as it builds on existing dynamic graph and anomaly detection techniques.
The paper tackles the problem of detecting novel and unseen network intrusions by proposing TGN-SVDD, a method based on dynamic graph modeling and deep anomaly detection, and demonstrates its superiority over baselines on realistic data.
With the growing digitalization all over the globe, the relevance of network security becomes increasingly important. Machine learning-based intrusion detection constitutes a promising approach for improving security, but it bears several challenges. These include the requirement to detect novel and unseen network events, as well as specific data properties, such as events over time together with the inherent graph structure of network communication. In this work, we propose a novel intrusion detection method, TGN-SVDD, which builds upon modern dynamic graph modelling and deep anomaly detection. We demonstrate its superiority over several baselines for realistic intrusion detection data and suggest a more challenging variant of the latter.