The amount of malware that target Android terminals is growing. Malware applications are distributed to Android terminals in the form of Android Packages (APKs), similar to other Android applications. Analyzing APKs may thus help identify malware. In this chapter, we describe how machine learning techniques can be used to identify Android malware. We begin by looking at the structure of an APK file and introduce techniques for identifying malware. We then describe how data can be collected and analyzed and then used to prepare a dataset. This is done by not only using permission requests and API calls, but also by using application clusters and descriptions as the source. To demonstrate the effectiveness of machine learning techniques for analyzing Android applications, we analyze the performance of support vector machine classification on our dataset and compare it to that of a scheme that does not utilize machine learning. We also evaluate the effectiveness of the features used and further improve the classification performance by removing irrelevant features. Finally, we address several issues and limitations on the use of machine learning techniques for analyzing Android applications.