Android Malware Analysis Using Ensemble Features

摘要: This paper presents a static feature extraction framework for Android malware analysis. The techniques are implemented by extracting prominent features from the components of application package i.e. AndroidManifest.XML files. Five different types likely permissions, count permission, hardware features, software as well API calls 1175 .apk files mined performing investigation. objective this work is to evaluate if independent effective in comparison ensemble features. Feature reduction performed investigate impact varied length on classification accuracy. selection such Bi–Normal Separation, Mutual Information, Relevancy score, Kolmogorov dependence and Kullback Leibler administered choose significant attributes. proposed method introduced here using dimensionality machine learning algorithms produces an overall accuracy 93.02% with Comparing empirical results individual former improved Separation.