A novel image based approach for mobile android malware detection and classification

Yapıcı M. M.

Knowledge-Based Systems, cilt.323, 2025 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 323
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1016/j.knosys.2025.113855
  • Dergi Adı: Knowledge-Based Systems
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC, Library and Information Science Abstracts, Library, Information Science & Technology Abstracts (LISTA)
  • Anahtar Kelimeler: Deep learning, Image processing, Malware detection
  • Ankara Üniversitesi Adresli: Evet

Özet

One of the most widely used mobile operating systems is Android. Consequently, it attracts the attention of hackers and is increasingly subjected to intensive attacks. To address this issue, this study proposes a image-based system for detecting Android malware and classifying malware families. The proposed approach has been tested separately on grayscale and RGB images. In similar studies within the literature, two fundamental issues have led to inconsistent and biased results. These issues are: 1) the problem of duplicate data within the datasets used, and 2) the problem of imbalanced data across classes. This study also offers solutions to these two issues. According to obtained results, it is observed that the proposed system achieved state-of-art results compared to previous studies, with an average accuracy of 0.987, precision of 0.987, recall of 0.986, and F1-score of 0.986.