User behavior feature extraction method, device, equipment and system

Abstract

The invention provides a user behavior feature extraction method, device, equipment and system. The method comprises the steps of collecting behavior data of a target user; inputting the behavior datainto a preset self-encoding network model for learning, wherein the preset self-encoding network model is obtained by training based on a preset neuron structure,the preset neuron structure comprisesa time interval sequence and an attention mechanism, the time interval sequence is used for reflecting the influence of an operation time interval on neuron memory updating, and the attention mechanism is used for reflecting the difference between a historical behavior and a current behavior; and taking a result output by a cognitive network in the preset self-encoding network model as behavior characteristic information of the target user. By utilizing the embodiment of the invention, the user behavior characteristics can be extracted more accurately, and the anti-fraud detection difficultyis reduced.

Description

technical field [0001] The present application relates to the technical field of feature mining, and in particular to a user behavior feature extraction method, device, equipment and system. Background technique [0002] At present, driven by emerging cutting-edge technologies such as big data, artificial intelligence, and blockchain, financial technology has become an irreversible trend to penetrate the traditional financial industry. At the same time, the risk of financial fraud is also expanding, and the anti-fraud situation is becoming more and more severe. [0003] Anti-fraud can be equated to predicting human behavior to a certain extent. At present, the industry's attempts to mine behavioral features mainly include artificially derived business features (feature engineering), expert experience, and supervised models. For example, in the anti-fraud business of financial institutions, more and more tend to dig deep into their own data, enrich user behavior data embedd...

AI Technical Summary

Problems solved by technology

However, as the forms of financial fraud continue to be updated and complicated, fraudsters continue to adopt new technologies to imitate the behavior trajectory of normal users. The unstructured and noisy characteristics of user behavior data make the mining of user behavior characteristics inefficient and difficult. Poor timeliness, which increases the difficulty of online anti-fraud detection

Images