Training method and device of service logic attack detection model, equipment and medium

By training the decision tree model and writing it to the Java virtual machine, the problem of lack of real-time business logic attack detection in the existing technology is solved, and accurate and rapid detection and classification of business logic attacks is realized, and the real-timeness of the detection model is improved. and accuracy.

CN118940045AActive Publication Date: 2024-11-12CHINA MOBILE GRP FUJIAN CO LTD +1
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411194181.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-12
Estimated Expiration
2044-08-28

AI Technical Summary

Technical Problem

In the prior art, business logic attack detection lacks real-time performance and cannot effectively detect and prevent cyber attacks using the expected functions and processes of the application in real-time.

Method used

A training method for business logic attack detection model is proposed. By obtaining the initial business data set and reference attack type, extracting detection indicators and feature values, training the decision tree model, and writing the model to a Java virtual machine to detect business logic attacks in real time .

Benefits of technology

It realizes accurate and fast detection and classification of business logic attacks, improves the training effect and real-timeness of the detection model, and can automatically trigger business logic attack detection when the Java virtual machine is started.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940045A_ABST
    Figure CN118940045A_ABST
Patent Text Reader

Abstract

The invention provides a service logic attack detection model training method and device, equipment and a medium. The method comprises the steps that an initial service data set and reference attack types are acquired, each reference attack type has at least one corresponding attack element, a detection index corresponding to each attack element is extracted from the initial service data set, and index feature values corresponding to the detection indexes are determined from the initial service data set; determining an annotation detection result corresponding to the initial service data set according to an index feature value corresponding to the detection index, and training an initial service logic attack detection model according to the initial service data set and the annotation detection result corresponding to the initial service data set to obtain a target service logic attack detection model, the model training effect of the service logic attack detection model can be effectively improved, and the target service logic attack detection model obtained by training can accurately and quickly detect and classify the already-arrived service logic attack.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method for constructing service logic attack detection model and computing device

    CN107920062A

  • Dynamic network abnormal attack detection method and device, electronic equipment and storage medium

    CN113206824A

  • Service logic vulnerability attack detection model training method and device

    CN116346456A

  • Model training method, evaluation method, device, equipment and medium

    CN117312861A

  • Theft attack detection method for image classification model

    CN117332409A