AndyWar: an intelligent android malware detection using machine learning

A personal Swiss army tool means a mobile device is now a necessity of every person of this world. Mobile devices are now a part of our daily life. And most of the mobile users are using android devices because of the flexibility and simplicity of android and also because of the huge open-source developer community. So the android supports more apps because of those developers. But as per Google Play Protect there were so many applications that do not follow the policy provided by them. So as a result Google deletes it from the play store. And few sneaky application developers are smart enough to escape those policies and successfully build a malicious app, that might be doing unusual things or may put viruses, worms, Trojans, malware, and many more. So in this paper, we proposed an android malware detection technique that helps to detect malware via its behavior using supervised learning. The proposed technique used a few well-known machine learning algorithms to detect and protect users from malware and data loss. We test our supervised model with various pre-existing datasets along with our one. After testing with various supervised models, our proposed voting algorithm achieves 97% of accuracy to detect malware. And we see that various malicious API calls and unusual behavior of the apps caught by our proposed system. So this proposed system may help researchers and users to understand the current situation with android apps and also provides a few approaches to develop future technologies for android systems.