To explore the efficiency of Cluster-GCN on Cybersecurity

Inhalt:
The detection and protection of Cyberattacks are always challenging due to the complexed and varied Cyberattack schemes. Research on knowledge graphs is becoming increasingly mature in many fields. To construct a cybersecurity knowledge base, researchers have combined the concept of the knowledge graph with cybersecurity. In this paper, we apply cluster-based graph convolutional network (Cluster-GCN) to explore the graph cluster structure, which emphasizes the harmful vulnerability and exposure for a system that providing a prevention of potential malicious attack. In order to explore the efficiency of Cluster-GCN on vulnerabilities detection, we also compare it with convolutional neural network and traditional graph convolutional network.