Android Malware Categorization and Characterization Using Machine Learning

Among all the current mobile operating systems, Android has the most market share with more than three billion active users. As android is open-source, it possesses many challenges of malicious cyberattacks. Most of the android applications are now online or somehow connected to internet and for this availability of connectivity with internet, hackers can push malware into the system easily without any concern of the user, although the application was safe and secure while installing for the first time. As a result, it is becoming difficult to identify their behavior and categorize to implement protection inside android operating system. In this research, we proposed a system to categorize and characterize various types of malware based on their network traffic pattern using K-means and Decision Tree algorithm, and also proposed a diagram demonstrating how android developers can implement this system in android OS.