Method and device for classifying network nodes

Abstract

Embodiments of the present invention provide a method and device for classifying network nodes. The method includes: obtaining the association information between nodes in the network to be classified and the category information of the nodes; Representation vectors of each node; using the acquired network representation learning model to classify the representation vectors of each node. The network node classification method and device provided by the embodiments of the present invention can fully encode the category information of the nodes into the final representation vector, thus effectively improving the classification effect on the node classification task.

Description

technical field [0001] The invention relates to the technical field of computer networks, in particular to a method and device for classifying network nodes. Background technique [0002] Network representation plays an important role in the field of network analysis. An effective network representation is of great help to network analysis tasks such as node classification, clustering, and link prediction. As the basic constituent elements of the network, nodes are generally represented as an independent discrete symbol, and this representation is called one-hot encoding representation. Because of the simplicity of this representation, it is widely used in the field of network analysis. However, this representation usually faces the sparsity problem and does not fully consider the relationship between nodes. [0003] Inspired by distributed representation learning in recent years, the concept of network representation learning is proposed to solve these problems. The pur...

AI Technical Summary

Problems solved by technology

Although the learned representations can be applied to different tasks, they may not perform well on specific prediction tasks

Images