Node embedding representation determination method and device, related equipment and program product

Abstract

The present disclosure provides a node embedding representation determination method and device, related equipment and program product, and relates to the technical field of computer and Internet. The method comprises: obtaining graph structure data, the graph structure data comprising a first node and a second node; obtaining pre-training embedding representations of the first node and the second node; processing the pre-training embedding representations of the first node and the second node through a neural network model to determine a predicted association relationship of the first node and the second node in the graph structure data; and training the neural network model through an actual association relationship and the predicted association relationship of the first node and the second node in the graph structure data, so as to determine embedding representations of each node in the graph structure data through the trained neural network model. The embodiment of the present disclosure can obtain the embedding representations of the nodes in the graph structure data through the method of supervised learning, and the embedding representations can accurately express the characteristics of the nodes in the graph structure data.