Amer, E., Mohamed, A., Mohamed, S., Ashaf, M., Ehab, A., Shereef, O., Metwaie, H. (2022). Using Machine Learning to Identify Android Malware Relying on API calling sequences and Permissions. Journal of Computing and Communication, 1(1), 38-47. doi: 10.21608/jocc.2022.218454
Eslam Amer; Ammar Mohamed; Seif ElDein Mohamed; Mostafa Ashaf; Amr Ehab; Omar Shereef; Haytham Metwaie. "Using Machine Learning to Identify Android Malware Relying on API calling sequences and Permissions". Journal of Computing and Communication, 1, 1, 2022, 38-47. doi: 10.21608/jocc.2022.218454
Amer, E., Mohamed, A., Mohamed, S., Ashaf, M., Ehab, A., Shereef, O., Metwaie, H. (2022). 'Using Machine Learning to Identify Android Malware Relying on API calling sequences and Permissions', Journal of Computing and Communication, 1(1), pp. 38-47. doi: 10.21608/jocc.2022.218454
Amer, E., Mohamed, A., Mohamed, S., Ashaf, M., Ehab, A., Shereef, O., Metwaie, H. Using Machine Learning to Identify Android Malware Relying on API calling sequences and Permissions. Journal of Computing and Communication, 2022; 1(1): 38-47. doi: 10.21608/jocc.2022.218454
Using Machine Learning to Identify Android Malware Relying on API calling sequences and Permissions
1Faculty of Computer Science - Misr International University - Cairo - Egypt
2Faculty of Graduate Studies for Statistical Research Cairo University
Abstract
The revolutionary in cyber attacks, especially in smartphones are rising. The Android operating system is becoming one of the most leading operating systems. Therefore, Android malware is rising in terms of popularity. Malware makers are using novel techniques to develop malicious Android applications, drastically diminishing the capabilities of traditional malware detectors. In consequence, those Anti-malware detectors become unable to detect these unexplained malicious apps. Currently, machine learning techniques are extensively used to discover new unknown Android viruses by analyzing the functionality of static and dynamic app reviews. In this paper, we introduce an Android malware detection technique based on API and permissions. Our purpose is to evaluate and examine the incorporation of machine learning classifiers with featured Android features such as APIs and permissions. We investigated several classification methods in characterizing Android malware with respect to the used feature. We discovered varied performance when we analyses all Android malware detection classifiers that use machine learning, suggesting that machine learning algorithms are effectively utilized in this area of identifying Android malicious apps.