Abnormal access behavior detection method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an abnormal access behavior detection method and device, electronic equipment and a storage medium. An abnormal destruction behavior of a malicious user can berecognized through access data of the user, and the characteristics that configuration is flexible, update is timely, and a recognition result is transparent and controllable are achieved. The detection method comprises the steps that an overall network structure is constructed according to historical user access data; current user access data is acquired; a current user access path is acquired according to the current user access data, then a corresponding skip path in the overall network structure is extracted according to the current user access path, and interface information contained inall pages of the skip path is acquired; and PV (Page View) data and click data of the current user are detected, whether the current user has a click behavior anomaly is judged, called interface information of the current user and the interface information contained in all the pages of the skip path are compared, and whether the current user has an interface calling behavior anomaly is judged.

Description

technical field [0001] The present invention relates to the technical field of the Internet, in particular to a detection method, device, electronic equipment and storage medium for abnormal access behavior. Background technique [0002] With the rise of Internet e-commerce, the number of online shopping users continues to increase. In order to gain user traffic and attention and expand the brand influence of their stores, online merchants often adopt some means such as promotions, snap-ups, and issuance of coupons. Correspondingly, these resources invested online are ultimately hoped to fall into the majority of ordinary users, bringing praise and transaction volume. Instead of being intercepted by some malicious users, a series of abnormal behaviors such as seizing coupons and preferential product quotas, and conducting secondary sales will affect the spread of activities and damage the interests of other users. [0003] Existing methods for detecting abnormal user access...

AI Technical Summary

Problems solved by technology

[0005] The method of building a model for classification by analyzing discrete feature points of historical users depends on the selection of feature sets and model construction, and the model does not have universal applicability; at present, the methods of swiping coupons and snapping up purchases through malicious means are flexible and changeable, and training models are used to There is a certain lag in the way of classification

Images