Automated feature selection using boruta algorithm to detect mobile malware

• Main Authors: Che Akmal, Che Yahaya, Ahmad Firdaus, Zainal Abidin, Salwana, Mohamad, Ernawan, Ferda, Mohd Faizal, Ab Razak
• Format: Article
• Language: English
• Published: The World Academy of Research in Science and Engineering 2020
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/43141/1/Automated%20Feature%20Selection%20using%20Boruta%20Algorithm%20to%20Detect%20Mobile%20Malware.pdf; http://umpir.ump.edu.my/id/eprint/43141/; https://doi.org/10.30534/ijatcse/2020/307952020

The usage of android system is rapidly growing in mobile devices. Android system might also incur severe different malware dangers and security threats such as infections, root exploit, Trojan, and worms. The malware has potential to compromise and steal the private data, classified data, instant messages, private business contacts, and confidential schedule. Malware detection is needed due to the malware continuously evolve rapidly. This research proposed automated feature selection using Boruta algorithm to detect the malware. The proposed method adopts machine learning prediction and optimizes the selecting features in order to reduce the model of machine learning complexity. Boruta algorithm is used to select features automatically for assisting the machine learning. The experimental results show that the proposed method is able to reach 99.73% accuracy in machine learning classification.