The multi-objective feature selection in Android malware detection system

Abstract

The Android operating system boosts a global market share over the previous years, which has made it the most popular operating system in the world. Recently, Android has become the target of attacks by cybercriminals because of its open-source code and its progressive growth. Many machine learning techniques have been used to address this issue in the Android operating system. However, a limited range of feature selection methods has been used in these systems. This paper, therefore, aims to address and evaluate the impact of a multi-objective feature selection approach called NSGAII in Android malware detection systems. To improve the diversity of solutions offered by this method, we have modified the standard NSGAII approach. Experimental results show that the proposed method can lead to better malware classification.