Network graph embedding method based on edges

Abstract

The invention discloses a network graph embedding method based on edges. The method comprises steps that an Edge2vec algorithm model is constructed; a random gradient descent method is utilized to train the Edge2vec algorithm model; the Edge2vec algorithm model is utilized to realize network graph embedding. The method is advantaged in that the edges of the network graph are directly mapped to thelow dimension vector space through a neural network model based on an automatic depth encoder, the local neighbor degree information and the global neighbor degree information between the edges can be reserved, compared with the prior art, the structural information of the edges of the network graph can be effectively reserved, better performance can be realized for network graph analysis tasks of the edges, and the method can be applied to the network graph analysis tasks of the edges.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an edge-based network graph embedding method. Background technique [0002] In recent years, network graph embedding has attracted the attention of more and more researchers. Network graph embedding studies how to effectively map a network graph to a low-dimensional vector space. This process is very helpful for analyzing network information. For example, we can apply the embedded graph to network graphs such as link prediction and node clustering. in the analysis task. [0003] The traditional network graph embedding method is mainly based on Laplace matrix or adjacency matrix for dimension reduction. However, the existing technology is often limited by computational overhead and performance. In recent years, some researchers have proposed a method based on the Skip-gram model. The principle of the network graph embedding method is to compare the nodes in the network graph t...

AI Technical Summary

Problems solved by technology

Therefore, the edge vector can only be obtained indirectly through the node vector, and the method of indirectly obtaining the edge vector will have performance defects, because using the vector of the end point of the edge to indirectly represent the edge vector cannot retain the complete information of an edge

Therefore, for some edge-based network graph analysis tasks such as link prediction, the method of indirectly obtaining edge vectors often cannot achieve optimal performance

Images