Overview of Mobile Attack Detection and Prevention Techniques Using Machine Learning

Authors

  • Ahmad K. Al Hwaitat The university of jordan https://orcid.org/0000-0003-4930-0074
  • Hussam N. Fakhouri Department of Data Science and Artificial Intelligence University of petra.
  • Moatsum Alawida Department of Computer Sciences, Abu Dhabi University, Abu Dhabi 59911, United Arab Emirates
  • Mohammed S Atoum King Abdullah II School of Information Technology, The University of Jordan, Amman 1142, Jordan
  • Bilal Abu-Salih
  • Imad K. M. Salah King Abdullah II School of Information Technology, The University of Jordan, Amman 1142, Jordan
  • Saleh Al-Sharaeh King Abdullah II School of Information Technology, The University of Jordan, Amman 11942, Jordan
  • Nabil Alassaf

DOI:

https://doi.org/10.3991/ijim.v18i10.46485

Keywords:

Mobile Security, Machine Learning in Cybersecurity, Intelligent Attack Detection, Adversarial Attacks and Defenses

Abstract


In light of the increasing sophistication and frequency of mobile attacks, there is a growing demand for advanced intelligent techniques capable of offering comprehensive mobile attack detection and prevention. This paper aims to critically evaluate the landscape of mobile security, outlining the evolution of mobile attack vectors and pinpointing the deficiencies in traditional security methods. The text embarks on a journey to understand the connection between machine learning (ML) and its promising applications in enhancing mobile security. First, we outline the current state of mobile attacks and the traditional methods used for their detection, emphasizing the clear limitations and the necessity for an innovative approach. Following this, we will elucidate the fundamentals of ML and its implications in cybersecurity, exploring the benefits it can provide to mobile attack detection frameworks. We delve into discussing various ML algorithms, such as decision trees, random forests, and support vector machines, highlighting their effectiveness and the metrics used to evaluate ML models in security tasks. Moreover, the paper sheds light on novel approaches such as semi-supervised and unsupervised learning in anomaly detection, as well as the applications of transfer learning in security. Addressing the pressing challenges faced in artificial intelligence (AI)-driven mobile attack detection, we delve deep into the intricacies of data collection, labeling, and the prevailing issues of imbalance and overfitting. Furthermore, we explore contemporary adversarial attacks and defenses, scrutinizing the real-world adaptability of AI models and the pivotal role of human-AI collaboration in enhancing attack detection mechanisms.

Downloads

Published

2024-05-22

How to Cite

Al Hwaitat, A. K., Fakhouri, H. N., Alawida, M., S Atoum, M., Abu-Salih, B., K. M. Salah, I., … Alassaf, N. (2024). Overview of Mobile Attack Detection and Prevention Techniques Using Machine Learning. International Journal of Interactive Mobile Technologies (iJIM), 18(10), pp. 125–157. https://doi.org/10.3991/ijim.v18i10.46485

Issue

Section

Papers