ABSTRACT
This chapter describes mobile app security analysis, one of the new emerging cybersecurity issues with rapidly increasing requirements introduced by the predominant use of mobile devices in people's daily lives. It discusses how big data techniques such as machine learning (ML) can be leveraged for security analysis of mobile applications. ML is a promising approach in triaging app security analysis, in which it can leverage the big datasets in the app markets to learn a classifier, incorporating multiple features to separate apps that are more likely to be malicious from the benign ones. The chapter also describes practice of employing a better evaluation strategy and better designs of future ML-based approaches for Android malware detection. It demonstrates the impact of some challenges on some existing ML-based approaches. The high imbalance in the positive and negative data samples in mobile data sets present unique challenges in both ML algorithm design and evaluation.
