Malware Detection in Android Mobile Devices by Applying Swarm Intelligence Optimization and Machine Learning for API Calls

Abstract

Attacks on mobile devices, such as smartphones and tablets, have been on the rise as their use has grown. Malware attacks are some of the most significant threats, resulting in a variety of security issues as well as financial losses. The feature space-restricted malware analysis helps to detect malware effectively. The purpose of this research is to find the most useful features of Application Programming Interface (API) calls to improve the detection accuracy of Android malware. Two Swarm Intelligence Optimization techniques, namely Bald Eagle Search (BES) & Sailfish Optimization (SFO) are evaluated with API Calls to identify the most promising features for Android Malware detection. The BES & SFO features selection techniques are assessed using machine learning classifiers such as K-Nearest Neighbour (KNN), Decision Tree (DT), Support Vector Machine (SVM), Linear Regression (LR) and Random Forest (RF). Experimentation resulted in an accuracy of 98.92% with 21 features out of 100 API call features.