A detection method and device for an Android malicious application

Abstract

The present invention provides a method and device for detecting Android malicious applications, which relate to the technical field of Android systems. The method includes acquiring a training sample set composed of malicious application samples and normal application samples; performing fusion feature extraction on the training sample set, and Select the extracted fusion features by principal component analysis to obtain Android application-related fusion features; establish a malicious application detection model based on extreme learning machine according to the relevant fusion features of Android applications; according to the malicious application based on extreme learning machine The detection model detects the Android application to be tested, and determines whether the Android application to be tested is a malicious application. The present invention can solve the problem that the current detection methods based on feature codes, static source code analysis or dynamic behavior may cause the detection of Android malicious applications to fail, and the malicious applications cannot be correctly determined.

Description

technical field [0001] The invention relates to the technical field of Android systems, in particular to a method and device for detecting malicious Android applications. Background technique [0002] Currently, with the rapid popularization of Android (Android) mobile smart terminals, the types and quantities of Android applications (APP) are also growing in a blowout. However, the ensuing network and information security issues are becoming more and more prominent, mobile smart terminal security incidents emerge in endlessly, mobile malicious applications are rampant, personal privacy theft, tariff consumption and other security issues occur from time to time, seriously affecting the healthy development of the industry. The above-mentioned problems are mostly caused by mobile malicious codes. Mobile malicious codes are usually implanted in the process of APP software development or secondary packaging, and grab economic benefits or spread spam through deception fraud, pri...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a detection method and device for an Android malicious application to solve the problem that current detection methods based on feature codes, static source code analysis, or dynamic behavior may cause the detection of Android malicious applications to fail and cannot be correctly determined. The problem with malicious apps

Images