基于统计学特征的android恶意应用检测方法

Android系统作为世界上最流行的智能手机系统，其用户正面临着来自恶意应用的诸多威胁。如何有效地检测Android恶意应用是非常严峻的问题。本文提出基于统计学特征的Android恶意应用检测方法。该方法收集5560个恶意应用和3000个良性应用的统计学特征作为训练数据集并采用聚类算法预处理恶意数据集以降低个体差异性对实验结果的影响。另一方面，该方法结合特征和多种机器学习算法（如线性回归、神经网络等）建立了检测模型。实验结果表明，该方法提供的两个模型在时间效率和检测精度上都明显优于对比模型。

Android malicious application detection method based on statistical features

Android system is the most popular smart phone system in the world and its users are facing many threats from malicious applications. How to effectively detect Android malicious applications is a very serious problem. This paper proposes an Android malicious application detection method based on statistical characteristics. This method collects statistical features from 5,560 malicious applications and 3,000 benign applications as training data sets and uses clustering algorithms to preprocess malicious data sets for reducing the impact of individual differences on experimental results. On the other hand, this method combines the features and various machine learning algorithms (such as linear regression, neural network, etc.) to establish a detection model. The experimental results show that the two models provided by this paper are superior to the comparison model in both time efficiency and detection accuracy.