Street-level IP address positioning method based on graph neural network

The invention provides a street-level IP address positioning method based on a graph neural network, and the method comprises the following steps: firstly, screening a landmark IP matched with a target IP; secondly, establishing a topological space relationship between a target IP and a landmark IP in the region from two different perspectives of IP features and network measurement so as to obtain an edge weight between the target IP and an associated landmark IP and form a complete topological graph structure; thirdly, utilizing an edge weight and a topological graph structure between the target IP and the landmark IP, using a graph convolutional neural network to aggregate neighborhood landmark features and geographic coordinates, obtaining probability representation about the target IP through regularization flow to resist negative effects brought by network fluctuation, and obtaining a robust hidden state about the IP; and finally, performing linear affine change by using a multi-layer perceptron, decodin