Mobile malware detection with machine learning: state of play and emerging challenges




Journal Title

Journal ISSN



Volume Title




Peer reviewed


Malware (or malicious software) on the mobile platform has been growing at an astonishing rate since smartphones started becoming popular over a decade ago. Android, the dominant mobile operating system worldwide, has been relentlessly targeted by malicious actors who have capitalized on its openness and global reach to monetize their malware. According to the 2019 Q1 McAfee Mobile Threat Report, more than 6 million new pieces of mobile malware were discovered in the wild in 2018 alone. Unfortunately, new malware is highly evasive to detection by traditional signature-based antivirus scanning. The shortcomings of the traditional anti-malware solutions have led to the emergence of research in machine learning (ML) based malware detection approaches in the last few years. Whilst the research results in this area has been encouraging, some challenges still remain. This talk will present the evolution of mobile malware and the state-of-the art research in machine learning based detection of Android malware. It will also discuss some emerging challenges and opportunities in the field of ML-based malware detection. The speaker will also present results of some of the recent research in ML-based malware detection that is being conducted at De Montfort University’s Cyber Technology Institute.



mobile malware, machine learning, concept drift, deep learning, attacks on machine learning models


Yerima, S. Y. (2019) Mobile malware detection with machine learning: state of play and emerging challenges. Cybersecurity Education and Research Conference (CERC2019), Kuwait, 11- 12 Nov. 2019.


Research Institute

Cyber Technology Institute (CTI)