With over 50 billion downloads and more than 1.3 million apps in the Googleofficial market, Android has continued to gain popularity amongst smartphoneusers worldwide. At the same time there has been a rise in malware targetingthe platform, with more recent strains employing highly sophisticated detectionavoidance techniques. As traditional signature based methods become less potentin detecting unknown malware, alternatives are needed for timely zero-daydiscovery. Thus this paper proposes an approach that utilizes ensemble learningfor Android malware detection. It combines advantages of static analysis withthe efficiency and performance of ensemble machine learning to improve Androidmalware detection accuracy. The machine learning models are built using a largerepository of malware samples and benign apps from a leading antivirus vendor.Experimental results and analysis presented shows that the proposed methodwhich uses a large feature space to leverage the power of ensemble learning iscapable of 97.3 to 99 percent detection accuracy with very low false positiverates.
展开▼
