Neural network system, method and device for carrying out risk evaluation on operation event

Abstract

The embodiment of the invention provides a neural network system and method for carrying out risk evaluation on an operation event. According to the scheme, firstly, an event sequence is obtained, theevent sequence comprises a plurality of operation events composed of a current operation event and historical operation events, and each operation event has multiple parts of corresponding attributeinformation; for each operation event, converting the multiple parts of attribute information of the operation event into multiple corresponding feature vectors, and performing first vector combination on the multiple feature vectors to obtain a first embedded vector corresponding to the operation event, the first vector combination comprising N-order inter-vector combination operation; and then performing second vector combination on the plurality of first embedded vectors corresponding to the plurality of operation events to obtain a second embedded vector corresponding to the event sequence. And processing the second embedded vector to obtain a processing vector, and determining a risk evaluation result of the current operation event at least according to the processing vector.

Description

technical field [0001] One or more embodiments of this specification relate to the fields of artificial intelligence and machine learning, and in particular to methods and devices for evaluating the risk of user operation events using a neural network system. Background technique [0002] With the rapid development of computer networks, network security issues have become increasingly prominent. There are a variety of high-risk operations, such as account theft, traffic attacks, fraudulent transactions, etc., which may threaten network security or user information security. For the sake of network security and risk prevention and control, in many scenarios, it is necessary to analyze and process user operation behavior or operation events, and evaluate the risk degree of user operation behavior for risk prevention and control. [0003] In order to assess the risk degree of an operation behavior, it can be analyzed based on the characteristics of the operation behavior itsel...

AI Technical Summary

Problems solved by technology

However, it is understandable that manual feature engineering consumes a lot of manpower and time, and the effect is heavily dependent on manual business experience and efficiency

When the business experience is not perfect, the features selected in artificial feature engineering may not be comprehensive or representative enough, making the accuracy of feature-based event analysis not high enough

Moreover, artificial feature engineering has the risk of security leakage

Images