Detecting Android Malware with an Enhanced Genetic Algorithm for Feature Selection and Machine Learning
B. Swarajya Lakshmi, Settipalli Naga Pranavi, C. Jayalakshmi, K. Gayatri, Moturi Sireesha, A. Akhila · International Journal of Research Publication and Reviews · 2023
Android's open source nature and Google's support have helped it garner the world's greatest market share.Being the most widely used OS in the world, it has attracted the focus of cybercriminals, who are active in a variety of ways but most notably via the widespread dissemination of malware software.In this research, we offer a machine-learning based method for Android malware detection that use an evolving Genetic algorithm to identify discriminating features.Machine learning classifiers are trained using the genetic algorithm's selected features and their ability to identify Malware is compared to its performance before and after feature selection.The experimental findings confirm that the genetic algorithm provides the best efficient feature subset, which aids in reducing the feature dimension to less than half of the original feature-set.For machine learning based classifiers, maintaining a classification accuracy of more than 94% after feature selection allows them to function on much decreased feature dimension, which has a beneficial effect on the computing cost of learning classifiers.