Android Botnets: A Proof-of-Concept Using Hybrid Analysis Approach

Ahmad Karim, Victor Chang, Ahmad Firdaus

Source Title: Journal of Organizational and End User Computing (JOEUC)32(3)

ISSN: 1546-2234|EISSN: 1546-5012|EISBN13: 9781522583707|DOI: 10.4018/JOEUC.2020070105

MLA

Karim, Ahmad, et al. "Android Botnets: A Proof-of-Concept Using Hybrid Analysis Approach." JOEUC vol.32, no.3 2020: pp.50-67. http://doi.org/10.4018/JOEUC.2020070105

APA

Karim, A., Chang, V., & Firdaus, A. (2020). Android Botnets: A Proof-of-Concept Using Hybrid Analysis Approach. Journal of Organizational and End User Computing (JOEUC), 32(3), 50-67. http://doi.org/10.4018/JOEUC.2020070105

Chicago

Karim, Ahmad, Victor Chang, and Ahmad Firdaus. "Android Botnets: A Proof-of-Concept Using Hybrid Analysis Approach," Journal of Organizational and End User Computing (JOEUC) 32, no.3: 50-67. http://doi.org/10.4018/JOEUC.2020070105

Export Reference

Favorite Full-Issue Download

View Full Text HTML

View Full Text PDF

Abstract

Mobile botnets are gaining popularity with the expressive demand of smartphone technologies. Similarly, the majority of mobile botnets are built on a popular open source OS, e.g., Android. A mobile botnet is a network of interconnected smartphone devices intended to expand malicious activities, for example; spam generation, remote access, information theft, etc., on a wide scale. To avoid this growing hazard, various approaches are proposed to detect, highlight and mark mobile malware applications using either static or dynamic analysis. However, few approaches in the literature are discussing mobile botnet in particular. In this article, the authors have proposed a hybrid analysis framework combining static and dynamic analysis as a proof of concept, to highlight and confirm botnet phenomena in Android-based mobile applications. The validation results affirm that machine learning approaches can classify the hybrid analysis model with high accuracy rate (98%) than classifying static or dynamic individually.