An optimization method and apparatus for policy configuration

Abstract

An optimization method and apparatus for policy configuration are disclosed. The method comprises: determining a policy configuration to be optimized in an initial policy; obtaining a prediction result outputted from the preset test data by using the current policy configuration in the policy, and performing statistics on the prediction result to obtain at least one index value, wherein the indexvalue is used for judging whether the current business index meets the preset requirements; If the current business indicator does not meet the preset requirements, optimizing the policy configuration, wherein the optimizing process comprises adjusting a combination mode of risk characteristics contained in the policy configuration, adjusting a numeric symbol used for representing the relationshipbetween the risk characteristics and the eigenvalue, or adjusting an eigenvalue corresponding to the risk characteristics contained in the policy configuration; After optimization, if the current business indicator meets the preset requirements, determining that the policy configuration in the current policy is available policy configuration.

Description

technical field [0001] The embodiments of this specification relate to the field of computer technology, and in particular, to a method and device for optimizing policy configuration. Background technique [0002] Third-party payment refers to an online payment model in which an independent institution with certain strength and reputation guarantee facilitates the transaction between the two parties through the connection with the Nets Union. In third-party payment, the user's operation behavior is recorded in the form of events, such as user login events, user registration events, user transaction events, etc., and the recorded events are sent to the risk control platform. The risk control platform utilizes The risk identification strategy and risk control strategy inside the platform identify and control the risks of events that represent user operation behaviors (also known as risk control events). If the events are risky, it means that the user's operation behavior has c...

AI Technical Summary

Problems solved by technology

[0004] At present, the way to optimize the strategy configuration in the risk identification strategy or the strategy configuration in the risk control strategy is: the operation and maintenance personnel analyze the existing cases based on their own experience (events that represent user operation behaviors are risky, but the risk control platform does not identify them. Events that show that there is a risk in the event that characterizes the user’s operation behavior, and the event that causes the user’s account funds to be damaged are called cases), and the strategy configuration in the risk identification strategy or the strategy configuration in the risk control strategy is optimized, that is, through manual methods. optimization

The existing method needs to rely on the experience of the operation and maintenance personnel, and the operation and maintenance personnel need to analyze the existing cases before they can optimize the policy configuration in the risk identification strategy or the strategy configuration in the risk control strategy, and the cycle is relatively short. long

Images