Generative adversarial network embedded representation learning method

Abstract

The invention relates to a generative adversarial network embedded learning representation method, which is applied to the technical field of network entity alignment. According to the invention, embedded representation of a network and cross-network entity alignment tasks are fused in a unified manner; network features are extracted through a graph convolutional neural network; meanwhile, generative adversarial learning is introduced to guide learning of domain invariant features, the influence of domain dependent features in the embedded representation learning process is avoided, on this basis, the graph convolutional network for sensing the direction is proposed to better optimize the structure information of the directed network, and based on the features of the graph convolutional network, the cross-network embedded representation learning efficiency is optimized through the weight sharing skill of the graph convolutional network. Compared with the prior art, the method effectively solves the problem that the alignment effect is poor due to the existence of domain features in the entity alignment tasks, obtains domain invariance features more beneficial to the entity alignment tasks through domain adversarial learning, and improves the entity alignment effect.

Description

technical field [0001] The invention relates to a generative confrontation network embedded representation learning method, in particular to an alignment task-oriented generative confrontation network embedded representation learning method, which is applied to the technical field of network entity alignment. Background technique [0002] The network entity alignment task was first applied in the field of bioinformatics, through the comparison of protein-protein interaction networks of different species, to find the common structure between proteins. Current entity alignment tasks are based on the same assumption that associated nodes should have a consistent connection structure across different networks. Due to the different functions and audiences of the network, there are often great differences between the network and the network. If a network is compared to a domain, the node attributes in the network often have a high degree of domain correlation, and suffer from the ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a generative adversarial network embedding for entity alignment to solve the problem that the existing entity alignment method based on embedded representation learning fails to fully consider the domain invariance characteristics. expression learning method

Images