Network training method, abnormal transaction behavior identification method and device, and medium

Abstract

The invention discloses a graph attention network training method, an Ethereum abnormal transaction behavior recognition method, a computer device and a storage medium. The training method comprises the steps of obtaining a first transaction network topological graph, feature information, a feature matrix and label information, training a graph attention network by using the first transaction network topological graph and the like. The trained graph neural network has the capability of capturing the correlation of each Ethereum address in the aspect of spatial information, can identify the behavior type corresponding to the label information at a relatively high identification rate, and has the capability of identifying the abnormal transaction behavior of the Ethereum; since the graph neural network performs feature information weighted summation on each node and the adjacent node thereof, the mode is only related to the adjacent node, the training process can be completed and the training effect can be achieved without using complete topological structure information of the first transaction network topological graph, and computer resources can be saved. The method is widely applied to the technical field of digital currency.

Description

technical field

[0001] The invention relates to the technical field of digital currency, in particular to a graph attention network training method, an abnormal transaction behavior identification method of Ethereum, a computer device and a storage medium. Background technique

[0002] Ethereum is a widely used blockchain platform with strong flexibility. Its token Ether is the digital currency with the largest market value in the world. Therefore, Ethereum has attracted the attention of many users, which makes Ethereum face a relatively large Great risk of being exploited by criminals. Criminals engage in illegal activities through digital currency technology, which usually produces abnormal transactions. Generally speaking, identifying abnormal transactions can provide stronger clues or evidence for criminals. Therefore, there is a problem of how to break through the Ethereum in the application of digital currency technology. The anonymity of digital currency technology i...

Images