IP street level positioning method and device based on graph neural network

The invention belongs to the technical field of target IP positioning, and discloses an IP street level positioning method and device based on a graph neural network, and the method comprises the steps: firstly, representing the traceroute original measurement data of a computer network as a graph with attributes; then, converting the graph with attributes into initial node embedding through an encoder; then, the initial node embedding is refined by modeling the connection information; finally, the decoder maps the fine embedding to the node position. According to the method, the convergence problem of GNN is relieved by considering priori knowledge, the geographic position prediction precision is improved, and experiments on different real data sets show that the method is superior to the most advanced rule-based and learning-based baseline by 16%-28% on the median error distance of all the data sets. 本发明属于目标IP定位技术领域，公开一种基于图神经网络的IP街道级定位方法及装置，该方法包括：首先，将一个计算机网络的traceroute原始测量数据表示成一个带属性图；然后，通过编码器将带属性图转换为初始节点嵌入；