Transaction risk control method and device

Abstract

The invention provides a transaction risk control method and device. The transaction risk control method in the invention comprises the steps of: receiving a to-be-judged order of an account, obtaining historical transaction data corresponding to the account, then, after the fact that pre-set transaction attributes need to be statistical according to the historical transaction data is judged, extracting transaction attribute data corresponding to the pre-set transaction attributes in the to-be-judged order and the historical transaction data, generating to-be-judged model characteristics according to the transaction attribute data, inputting the to-be-judged model characteristics into a pre-set risk control model, so that a risk result is obtained, and finally, performing risk control on the to-be-judged order according to the risk result. According to the transaction risk control method and device provided by the invention, because the pre-set transaction attributes are transaction attributes that the difference value among corresponding transaction attribute data in a normal order and a fraud order is greater than a pre-set numerical value in a process of training the pre-set risk control model, the transaction attribute data input into the model can comprehensively reflect the to-be-judged order; and thus, the judgement accuracy of the risk control model is increased.

Description

technical field [0001] The present invention relates to the field of transaction security, in particular to a transaction risk control method and device. Background technique [0002] With the development of the Internet, Internet business is becoming more and more abundant. The subsequent Internet business fraud is also increasing. Therefore, in order to ensure the security of information operations, usually the system to which an Internet business belongs needs to have a risk control system, and the core of this system is the risk control model. The risk control model can be obtained by training the relevant transaction data in the completed order records as input. With the help of the risk control model, the risk identification of the received order can be carried out through the model. [0003] However, when the transaction information data on the order becomes more and more complex, most of the input data types of the model are not well targeted, resulting in a high ...

AI Technical Summary

Problems solved by technology

[0003] However, when the transaction information data on the order becomes more and more complex, the types of input data o

Images