Evolving trees for detecting android malware using evolutionary learning

Document Type : Research Paper


Faculty of Computer Science and Mathematics, University of Kufa, Iraq


Android smartphone platforms have grown rapidly and become ubiquitous due to the fact that they have opened up new possibilities in every aspect of modern life. However, as a consequence of their widespread, such systems suffer an increasing frequency of malware attacks. New malware or modifications to existing malware are increasingly being used by attackers in such operations. Security researchers deal with a continually evolving threat environment that is broad, complicated, and ever-changing. Conventional detection methods failed to adequately protect such systems and so the security industry has started to detect systems that use machine learning techniques are becoming more common. In this work, we used an evolutionary algorithm to identify malware targeting Android smartphones. In order to evaluate our system, we compared it to different state-of-the-art algorithms. Finally, we demonstrated our proposed method's ability to detect zero-day malware. Obtained results show that our method achieved 99.11\% detection accuracy, and is well-suited for detecting zero-day malware.


Volume 14, Issue 1
January 2023
Pages 753-761
  • Receive Date: 19 April 2022
  • Revise Date: 02 May 2022
  • Accept Date: 16 July 2022