COVID-Themed Android Malware Analysis and Detection Framework Based on Permissions

In the context of the COVID-19 pandemic the malicious actors actively creating COVID-themed android malicious apps and without much attention user may often grant all the required permissions to install those fake apps. The Android permissions are crucial sources of vulnerability. This vulnerability often leads to major privacy threats. In this work COVID-themed android malwares were collected and analyzed to develop a detection framework based on the static feature permission and machine learning techniques. The proposed system analyses 100 COVID-themed fake applications which released in 2020. The sensitive permissions are selected using Recursive Feature Elimination (RFE) technique. The study shows better accuracy of 0.830 and 0.812 with Decision tree classifier and Random forest classifier respectively.