科技创新与应用2025，Vol.15Issue(18)：45-49,5.DOI:10.19981/j.CN23-1581/G3.2025.18.010
基于集成学习的安卓恶意软件特征提取与检测方法
冯志峰1
作者信息
- 1. 宁波城市职业技术学院,浙江 宁波 315000
- 折叠
摘要
Abstract
Android,as the most popular operating system today,offers convenience to users through its openness and wide application.However,this same openness also provides opportunities for malware development,posing significant threats to users'personal privacy and data security.To address this issue,this study proposes an integrated learning-based method for feature extraction and detection of Android malware.The authorization request of Android APK is extracted as feature points through automated scripts,combined with an enhanced support vector machine(E-SVM)model and a convolutional neural network(CNN)model for integrated learning training,generated a hybrid model,and used to improve the detection rate of Android malware.Final experimental data shows that the detection accuracy rate for malware reaches more than 96%.关键词
恶意软件/机器学习/深度学习/继承学习/特征提取检测Key words
malware/machine learning/deep learning/inheritance learning/feature extraction and detection分类
信息技术与安全科学引用本文复制引用
冯志峰..基于集成学习的安卓恶意软件特征提取与检测方法[J].科技创新与应用,2025,15(18):45-49,5.
