A kind of risk identification method and device

Abstract

The present application discloses a risk identification method and device, wherein the method includes: identifying the current event as a risk event, searching for historical events associated with the risk event based on event attributes, and updating the risk identification of the historical event from a non-risk event to Risk events, so that for the related historical events before this risk event, consider the possible risks, increase the corresponding mechanism for backtracking the risks of historical events, and avoid missing risks that have not been identified before due to various reasons. The ability to identify historical event risks has been improved.

Description

technical field [0001] The present application relates to the technical field of network security, in particular to a risk identification method and a risk identification device. Background technique [0002] As network services cover more and more fields, the security of network services becomes more and more important. While online business brings convenience to people's life, it also brings risks, especially in the fields of e-commerce and online finance. It is very important to effectively control the risk of online business. [0003] Risk control refers to the safeguard measures and methods for Internet business security, which eliminate or reduce the various possibilities of risk events, or reduce the losses caused by risk events. Business behavior events that usually require risk prevention and control are called risk control events, such as logins and transactions. When a user generates a risk control event in its own business or a third-party business, it needs to...

AI Technical Summary

Problems solved by technology

[0004] When the existing risk control engine conducts risk identification, it ignores the possible risks of the user's event behavior before this risk behavior, lacks a mechanism for backtracking on event risks, and easily misses the risk that has not been identified before due to various reasons. The identified risks, the identification ability of risk identification is not enough

Secondly, the risk control engine needs to obtain a large amount of data in real time to identify risks. In scenarios with high performance requirements, such as the scenario of millions of concurrent requests during promotional activities, it takes a long time to obtain a large amount of data in real time, which directly affects risk control. The performance of the engine cannot guarantee millisecond-level services

Thirdly, due to its own problems, the risk control engine cannot normally identify risk control events, resulting in a large number of wrong risk identification results, which has a negative impact on the normal operation of network services

Images