An Assessment of the Recognition and Identification of Malware for Android Employing Optimized Choice of Features Based on Genetic Algorithms

The Android platform is the biggest global market distributes since of its open source nature and support from Google. Due to the widespread dissemination of malicious programs, the majority extensively worn operating system in the world has attracted the attention of cybercriminals. This investigation suggests an efficient machine-learning method for Android malware detection that uses an evolutionary genetic algorithm to choose features derived from discrimination. Machine learning classifiers are trained using exact features by means of genetic algorithms, and their ability to identify malware prior to and following choosing features is evaluated. The experimentation findings confirm that the best optimized feature subset produced by the genetic algorithm aids in reducing the dimension of the feature to fewer than half of the initial feature set. Machine learning-based classifiers sustain a categorization exactness of over 94.2% after feature selection, even with a substantially decreased feature dimension. This absolutely affects the computing complication of learning classifiers.