| Title |
Android Malware Detection System using Deep Learning and Code Item |
| Authors |
(Seung-Pil W. Coleman) ; (Young-Sup Hwang) |
| DOI |
https://doi.org/10.5573/IEIESPC.2021.10.2.116 |
| Keywords |
Android malware detection; Code item; Convolutional neural network; Grayscale image; Static analysis |
| Abstract |
This paper proposes an Android malware detection method that reduces the overhead of 2-dimensional image generation from Android packages (APK) to build deep learning models that effectively discern whether an application is malware. Other image-based malware detection methods typically use the whole Android application executable file (DEX file) or a large section that often contains redundant information. However, our technique generates grayscale images using minimal representative data from the code item section. Two-dimensional images are utilized by a state-of-the-art feature extractor and spatial pattern recognition technique with a convolutional neural networks (CNN) architecture for image classification. Positive results were obtained for the execution time and memory usage compared to other methods. The code item section binaries contain relevant information about an Android application. |