Artificial Intelligence based Algorithm for Detecting Android Obfuscated Applications | ||||
International Journal of Intelligent Computing and Information Sciences | ||||
Volume 24, Issue 1, March 2024, Page 1-11 PDF (571.22 K) | ||||
Document Type: Original Article | ||||
DOI: 10.21608/ijicis.2024.250295.1308 | ||||
View on SCiNiTO | ||||
Authors | ||||
Hend Faisal Aboud 1; Hanan Hindy2; Samir Gaber3; Abdel-Badeeh M. Salem 4 | ||||
1Computer science, Ainshams University, Egypt | ||||
2Faculty of Computer and Information Sciences, Ain Shams University, Egypt | ||||
3Faculty of Engineering in Helwan, Helwan University | ||||
4Computer Sciece Department, Faculty of Computer and Information Sciences, Ain Shams University | ||||
Abstract | ||||
As technology continues to advance, so does the landscape of Android; based on its open-source nature which renders it vulnerable to various risks. Therefore, the developers need to deploy and employ obfuscation techniques in their newly developed android applications. In this paper , we present an investigation into Android obfuscation detection. Our work encompasses a comprehensive examination of Android obfuscation techniques and an exploration of their intersection with machine learning. We conducted extensive experiments involving various machine learning models to detect obfuscation. Among these models , The results show that Random Forest is the one with the most promising results with accuracy 99.5% in detecting Android Obfuscation. The dataset utilized in the experiments encompasses a diverse range of samples, including both malicious and benign samples. This diversity allows for a robust evaluation of the effectiveness of obfuscation detection across different scenarios and highlights the challenges posed by varying obfuscation techniques. | ||||
Keywords | ||||
Android Obfuscation; Machine Learning; Artificial Intelligence; Information Security | ||||
Statistics Article View: 46 PDF Download: 51 |
||||