计算机与数字工程2019，Vol.47Issue(5)：1184-1188,5.DOI:10.3969/j.issn.1672-9722.2019.05.033
基于Stacking的Android恶意检测方法研究
Malware Detection System of Android Applications Based on Stacking
董克源 1徐建1
作者信息
- 1. 南京理工大学计算机科学与工程学院 南京 210094
- 折叠
摘要
Abstract
For the weakness of single data mining algorithm for Android malicious application detection,this paper proposes an Android malicious detection method based on Stacking. Targeted at the Android applications as the research objects,the im?proved feature extraction methods are used to extraction permission feature and train base classification models. Finally,a new classi?fication model is trained based on the result of each classification model using the idea of ensemble learning. The experimental result and analysis of real applications in application market show that the Android malicious detection method based on Stacking can im?prove the detection accuracy of malicious applications.关键词
Android/恶意检测/权限特征/分类/Stacking方法Key words
Android/malware detection/permission feature/classify/Stacking method分类
信息技术与安全科学引用本文复制引用
董克源,徐建..基于Stacking的Android恶意检测方法研究[J].计算机与数字工程,2019,47(5):1184-1188,5.
