Network payment anti-fraud system architecture design method based on suspected fraud transaction reference order

Abstract

The invention discloses a network payment anti-fraud system architecture based on suspected fraud transaction reference order, and relates to the field of network payment anti-fraud. According to themethod, a fraud feature subspace synthesis concept is creatively adopted, a complete fraud space division method based on a suspected fraud transaction reference order is provided for the first time,and a fraud feature space is divided into a fraud transaction predecessor feature space, a fraud transaction post-secondary feature space and a fraud transaction median feature space. Different feature subspaces refer to the order according to the transaction data generated in the transaction process of the user, several or all of the modules are used for analyzing whether the transaction is a fraudulent transaction or not, the results of the modules are integrated through the decision strategy of the voting method according to the results generated by the modules, and finally the decision result of the system is generated. According to the invention, the coupling property is reduced by utilizing the mutual independence of multiple modules, the multiple modules are combined by utilizing anintegration and coordination strategy, the expansibility and interpretability are good, the functions of the modules are complementary, and the anti-fraud system performance is improved.

Description

technical field [0001] The invention relates to the field of network payment anti-fraud, in particular to the architecture design of a network payment anti-fraud system. Background technique [0002] With the development of technology and service models such as mobile Internet and e-commerce, online payment has penetrated into people's daily life. Online payment can be used not only for online shopping, but also for fast offline payment, which greatly facilitates people's lives and improves efficiency. While online payment brings certain convenience, it also brings new risks, such as fraudulent issues such as fraudulent brushing and money laundering caused by identity theft or loss of mobile phones. For many online trading platforms, when fraudulent gangs break through the anti-fraud algorithm, they can cause huge economic losses (usually measured by the loss rate), so the anti-fraud work for online payment is particularly important. [0003] Most of the existing network p...

AI Technical Summary

Problems solved by technology

[0004] Traditional financial anti-fraud methods are difficult to meet the needs of online payment anti-fraud field

Images