What is it about?
In this paper, we propose a framework to detect cyberattacks in critical infrastructures. This framework includes two phases: 1) Multi-view Causal Graphs and Spectral Fusion, where we learn the dense view and sparse view causal graphs from sensory data streams and fuse the two causal graphs into a single weighted Laplacian matrix representation. 2) Graph Anomaly Detection, where we train a Deep Convolutional Graph Neural Network (DGCNN) on the Laplacian representation of the graphs to detect attack statuses on sensory data streams per time interval.
Featured Image
Why is it important?
This is one of the first works that delve into exploring the power of Causal Graph Learning to enhance cyberattack detection.
Perspectives
In this paper, we incorporate causal graphs to capture invariant anomaly patterns in varying streams of real-time data; and then introduce multi-view fusion for robust attack pattern representation to detect cyberattacks efficiently.
Arun Vignesh Malarkkan
Arizona State University
Read the Original
This page is a summary of: Multi-view Causal Graph Fusion Based Anomaly Detection in Cyber-Physical Infrastructures, October 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3627673.3680096.
You can read the full text:
Contributors
The following have contributed to this page
