Anti-fraud method and device based on historical data playback

Abstract

The embodiment of the invention provides an anti-fraud method and device based on historical data playback. The method includes: synchronizing user history data; Obtaining the risk transaction time and transaction type set by the user; Judging whether a risk analysis can be performed according to the user history data, the risk transaction time and the transaction type; If it is judged that the risk analysis can be performed,analzying the transaction data within the risk transaction time, and combining the user history data and the data generated by the production system. The anti-fraud methodand the device based on historical data playback provided by the embodiment of the invention realize the system hot start function by retracing the historical data, can bring the historical businessdata into the statistical system, realize the linkage combination of the historical data and the real-time data, comprehensively mine the value of the historical data, and realize the system hot start.

Description

technical field [0001] The embodiment of the present invention relates to the field of computer technology, in particular to an anti-fraud method and device based on historical data playback. Background technique [0002] As an excellent fraud risk identification tool, the real-time anti-fraud rule function is being adopted by more and more financial companies. After years of development, the anti-fraud rule function has gradually formed a complete system. [0003] In the existing technology, the system implementation mechanism of most rule functions, the statistics of the new configuration of the system cannot count the historical data that has occurred, that is to say, the existing anti-fraud system is a cold start when it goes online again. For example, "the number of login times of the user's direct sales bank on the same device in the past 6 months", because historical data cannot be used, the customer's data deviates from the actual data in the previous 6 months. [0...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, the system implementation mechanism of most rule functions, the statistics of the new configuration of the system cannot count the historical data that has occurred, that is to say, the existing anti-fraud system is a cold start when it goes online again. For example, "the number of login times of the user's direct sales bank on the same device in the past 6 months", because the historical data cannot be used, the customer's data in the first 6 months deviates from the actual

[0004] Therefore, the statistics of the longer time window and the rules based on these statistics deviate greatly from the actual situation, which will affect the application effect of the anti-fraud system in the initial stage of production

Images