基于机器学习的Android恶意软件检测实验

Android malware detection experiment based on machine learning

针对传统Android恶意软件检测技术准确率不足的问题,该文设计并用实验验证了一种基于机器学习的Android恶意软件检测方法。首先提取Android应用的一些关键静态特征,生成对应的特征向量,然后采用机器学习算法构建并训练模型,最后对待测Android应用软件进行分类检测。实验结果表明,在使用相同数据集的情况下,支持向量机(SVM)算法取得了96.4%的检测率,略高于逻辑回归、朴素贝叶斯、决策树算法。通过实验,不仅使学生对Android应用软件有了更清晰的认识,还提升了学生用机器学习方法分析和解决实际问题的能力。

In view of the problem that the detection accuracy of traditional Android malware detection technology is insufficient,a machine learning based Android malware detection method is designed and tested.Firstly,the key static features of Android applications are extracted,and the corresponding feature vectors are generated.Then,the model is trained by machine learning algorithm.Finally,the Android applications are classified and detected.Experimental results show that with the same data set,the detection rate of SVM is 96.4%,which is slightly higher than the logistic regression,naive Bayes and decision tree algorithm.Through the experiment,it not only enables students to get a clearer understanding of Android application software,but also improves their ability to analyze and solve practical problems with machine learning methods.