Android Malware Detection Using Genetic-Algorithm-based Feature Selection and Machine Learning

Android Malware Detection Using Genetic-Algorithm-based Feature Selection and Machine Learning

GAMIDI NAGA VENKATA SIVARAMAKISHORE1

1UG Scholar,Department of  Electronics and communication Engineering,SRM university

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - The Android platform has the largest global market share because of its open-source nature and Google's support. Windows has grabbed the attention of fraudsters who exploit it to distribute malware since it is the most commonly used operating system on the planet. To produce the best-optimized feature subset that may be utilized to train machine learning algorithms most efficiently, the suggested methodology uses an evolutionary Genetic Algorithm for discriminatory feature selection. The ability of machine learning classifiers to detect malware before and after feature selection is examined. According to the outcomes of the experiments, the genetic algorithm gives the ideal feature subset, lowering the feature dimension  to less than half of its original feature set. Machine learning-based classifiers maintain good classification accuracy despite working with lower and smaller feature dimensions, which reduces computational complexity.

Key Words:  Genetic algorithm, Machine learning, Android malware, Reverse engineering