Research on Industrial Network Worm Propagation Traceability Based on Improved GCN

With the increasing number of industrial network worm attacks, traceability technology plays an increasingly important role in network security. Aiming at the problems of low accuracy and complex computation in traditional traceability methods, this paper proposes several methods of network worm propagation traceability based on improved Graph Convolutional Networks (GCN). Compared with traditional traceability methods, the proposed tracing method can predict the propagation source nodes more effectively and accurately by combining graph convolutional neural network. By comparing the performance of the original GCN and the improved GCN (including the optimization of graph attention mechanism), the experimental results show that the improved GCN has better accuracy than the traditional algorithm and the original GCN in most cases.