Machine Learning Approaches for Detecting Malware in Android Applications

Abstract

The rapid growth of Android applications has significantly increased the risk of malware attacks, posing serious threats to user privacy and system security. This study presents a machine learning-based approach for Android malware detection using optimized feature selection techniques. A dataset containing application features is preprocessed and divided into training and testing sets. Traditional classifiers such as Support Vector Machine (SVM) and Artificial Neural Networks (ANN) are implemented to identify malicious patterns. To enhance detection performance, a Genetic Algorithm (GA) is employed for feature selection, reducing dimensionality and improving model efficiency. The optimized models, namely SVM with GA and ANN with GA, are evaluated using performance metrics such as accuracy, classification report, and confusion matrix. Experimental results demonstrate that GA-based feature selection significantly improves classification accuracy and reduces computational time compared to conventional methods. The proposed system provides an effective and scalable solution for detecting Android malware in real-world scenarios.