Method, device and apparatus for detecting malicious application and training detection model

Abstract

A method for training a malicious application detection model includes extracting sample features based on an application sample with a sample tag for marking a security state of the application sample, Using the sample tag and the extracted sample features to obtain a first non-linear relationship model between the sample features and a security state by training and output a first prediction result; And obtaining a second non-linear relationship model between the sample characteristics and the safety state by training according to the corrected first prediction result and the sample characteristics of the partial application to be detected when the first prediction result corresponding to the partial application to be detected is determined to be wrong. The scheme based on the embodimentof the invention can improve the coverage rate and the accuracy rate of malicious application detection by training the detection model with the labeled application sample and training the supplementary model with the predictive error sample combined with feedback.

Description

Technical field [0001] The present invention relates to the technical field of computer applications, in particular to a method, a device, a computer-readable storage medium and an electronic device for detecting malicious applications and training detection models. Background technique [0002] In recent years, with the rapid development of the Internet, especially the mobile Internet, more and more service providers choose to provide paid or free services to users through applications (App). Correspondingly, the number of apps installed in user terminals (such as mobile devices or personal computers) is increasing. While this provides great convenience to users, it also leaves room for malicious applications to survive. [0003] A malicious application usually refers to an application that is developed by a developer by inserting code that endangers the safety of the user's interests in the application, which causes loss of profit and security risks to the user. Possible harmfu...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem in the related art that malicious application detection is likely to cause manslaughter or missed killing, the present invention provides a method, device, computer-readable storage medium and electronic equipment for detecting malicious applications and training detection models

Images