User request detection method and device, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a user request detection method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining equipment data of a terminal sending a user request; constructing a feature set by adopting a preset score feature pair combined features extracted from the equipment data; inputting the feature set into a plurality of anomaly detection models according to the type of the feature set, obtaining a detection result for judging whether the user request is abnormal or not; and judging the detection result by adopting a preset judgment method so as to determine whether the user request is abnormal or not, and training the abnormal detection model to a convergence state by adopting a positive sample or negative sample feature set in advance, thereby carrying out security classification on the terminal through the feature set. According to the method, a feature set is constructed for text type equipment data with nominal attributes, an effective classification feature set can be mined, and the recognition accuracy is improved.

Description

technical field [0001] The embodiments of the present invention relate to the financial field, in particular, a method, device, computer equipment, and storage medium for detecting user requests. Background technique [0002] With the development of Internet technology, the network plays an increasingly important role in people's work, life and study. In order to ensure the user's network security, it is necessary to detect whether there is any abnormality in the network user from time to time. Abnormal behavior refers to the behavior that affects the normal operation of the network. [0003] Usually, when detecting whether there is an anomaly in a user, the result is usually inaccurate due to a poor detection model. For example, when sending a user registration request or a verification request, the server obtains the device data of the terminal as sample data, and uses the sample data to build a model. However, since the data of the eye version are basically nominal att...

AI Technical Summary

Problems solved by technology

However, since the data of the eye version are basically nominal attributes, only a small number of data are numerical. For the data of nominal attributes, it is difficult to mine effective classification features, resulting in poor models, which in turn reduces the accuracy of detecting abnormal behaviors.

Images