Deep-Learning Based Approaches to Android Malware Classification
Mobile devices are becoming more and more popular as people start seeing them as a requirement for school, work, and every aspect of life. These devices allow us to check our social media, talk to our friends and coworkers, and even interact with our bank accounts. Since these devices have control over so much about our lives, we want to make sure that the phone owner is the only one who has access to the information on the phone.
Many attempts have been made to secure Android devices from the Google Watchdog program that currently watches what you put in the Google Play store to the anti-virus programs on the phone. However, these measures are not enough to prevent malware from being distributed to Android phones. There are certain types of malware that these tools just don't catch.
Our research focuses on trying to identify malicious programs on the Android devices and the family for the malware. Each piece of malware fits into a family based on its behavior and there are hundreds of families of malware in the Android community. In the field of malware detection, there are some malware families that no one is able to detect. Our goal is to make a program that focuses on detecting the types of malware families that no other programs are able to detect currently.