Detection of Android Malware by Using Classification Algorithms

Abstract

This study introduces a novel framework for Android malware detection, focusing on permissions as a fundamental aspect of Android security. Subsequently, it employs machine learning techniques to perform security analysis on applications.Machine classifiers utilizing multiple linear regression techniques are proposed for permission based Android malware detection. These classifiers are subjected to comparison against fundamental machine learning algorithms, including MLPClassifier, LinearDiscriminantAnalysis, RandomForestClassifier and LinearRegression are used for detecting android malware files. Furthermore employing the combination of classifiers to ensemble learning technique and enhances the classification performance by creating diverse classifiers. The study demonstrates remarkable performance using classification algorithms grounded using MLPClassifier, LinearDiscriminantAnalysis, RandomForestClassifier and LinearRegression models for obviating the necessity for overly existing techniques to show moure accuracy.