Abnormal node recognition method and device, model training method and device and storage medium

Abstract

The invention provides an abnormal node recognition method, an abnormal node recognition model training method, an abnormal node recognition model training device, an abnormal node recognition model training system and a storage medium, the abnormal node recognition model training method comprises the following steps: constructing an explicit association graph based on an explicit association relationship between transaction elements, the transaction elements at least comprising a transaction account; constructing an implicit association graph based on the similarity among the plurality of transaction accounts, wherein the similarity among the plurality of transaction accounts is calculated according to the historical transaction data of the plurality of transaction accounts; fusing the explicit association graph and the implicit association graph to obtain an association graph taking the transaction account as a node; and training a graph neural network based on the features and labels of each transaction account and the association graph to obtain an abnormal node recognition model. By means of the method, the complete and high-accuracy association graph structure can be constructed, and the recognition effect of the node recognition model is improved.

Description

technical field [0001] The invention belongs to the field of node identification, and in particular relates to an abnormal node identification method, a model training method, a device and a storage medium. Background technique [0002] This section is intended to provide a background or context for implementations of the invention that are recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] In recent years, there have been a large number of fraudulent transactions in online transaction payments, such as counterfeit registration transactions, which have brought considerable losses to users. In order to improve the level of preventing fraudulent transaction risks in online transaction payments, efficient and automatic detection of counterfeit registered transaction cards has been carried out, and risk disposal of suspected counterfeit registered bank cards output by the model has been carried out. [0004] I...

AI Technical Summary

Problems solved by technology

However, in reality, due to the influence of missing information and noisy connections, the edges of the constructed association network are likely to be incomplete or not all real.

This can easily lead to graph computing methods obtaining unsatisfactory node representations on the wrong graph structure, which affects subsequent downstream tasks

[0005] Therefore, how to construct a complete and correct graph structure for subsequent association graph analysis still faces certain challenges

Images