High-order neighborhood hybrid network representation learning method and device

Abstract

The invention discloses a high-order neighborhood hybrid network representation learning method and a device. A self-attention mechanism and a cascade aggregation layer are added on the basis of an original graph convolution layer, and the method comprises the following steps: converting a Laplace matrix of a graph into a node pair graph attention matrix by using the self-attention mechanism, andtraining weight parameters to learn different attention coefficients; gathering information flows with different distances through the cascade aggregation layer, and taking the output of the previousorder as the input of the next order to control the calculation complexity; and determining that the embedded vector is output to a downstream machine learning task, or outputting a classification result. According to the method, real end-to-end training can be realized, the training speed of the model is effectively improved, and the proposed idea of network high-order and low-order information hybrid learning has field expandability and is simple and easy to realize.

Description

technical field [0001] The present invention relates to the field of data mining and network technology, in particular to a method and device for learning network representation of high-order neighborhood mixing. Background technique [0002] At present, the mainstream network representation learning system in the industry is mainly designed based on the traditional pipeline method. The graph corpus is generated by random walk, input to the natural language model to train the node embedding vector, and then output to the downstream machine learning task to complete the final result. Target. This method involves mutual coordination between multiple modules, and feedback from non-end-to-end users is difficult to upload back to the upstream module, and cannot directly affect the learning of the task module. For example, a social network representation method based on bidirectional distance network embedding mainly goes through three steps: constructing node unique codes, gener...

AI Technical Summary

Problems solved by technology

Although this method considers the diversity of node-to-relationships, because it is based on the exploration of existing relationships, it does not fully consider the interaction between high-order nodes, and does not really model the relationship between nodes in the entire network.

Images