DeepMalOb: Deep Detection of Obfuscated Android Malware

Publication Type

Book Chapter

Publication Date (Issue Year)

2023

Journal Name

Pan-African Artificial Intelligence and Smart Systems. PAAISS 2022. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering. Springer, Cham.

Abstract

The detection of malware android became very crucial with the use of obfuscation techniques by developers of malicious applications. In the literature several approaches have been proposed to take into account certain techniques. But it is difficult to take into account all obfuscation techniques because of mutations and this is a critical challenge for cybersecurity. In this contribution, we proposed an approach to detect obfuscated malicious applications. This approach is based on the memory dump process. This process helps to discover the behaviour of obfuscated applications while they are executing without targeting a particular obfuscation technique. We implemented our application using supervised neural networks. We tested and selected hyper-parameters to train our detection model. The different results obtained by the evaluation metrics such as accuracy, precision, recall and F1 score, are excellent with high values around 99%.

Keywords

DeepMalOb, Deep Detection, Obfuscated, Android Malware

Rsif Scholar Name

Zakaria Sawadogo

Rsif Scholar Nationality

Burkina Faso

Cohort

Cohort 2

Thematic Area

ICTs Including Big Data and Artificial Intelligence

Africa Host University (AHU)

University of Gaston Berger (UGB), Senegal

Share

COinS