A Novel Mobile Malware Detection Model Based on Ensemble Learning

Karakaya A., ULU A.

2023 Innovations in Intelligent Systems and Applications Conference, ASYU 2023, Sivas, Türkiye, 11 - 13 Ekim 2023 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/asyu58738.2023.10296543
  • Basıldığı Şehir: Sivas
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: blockchain, cryptography, information security, post-quantum
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Mobile device usage is rapidly increasing with the popularity of smartphones. This situation makes mobile devices a target for cyber attacks. Previously, personal computers or server systems were more commonly targeted by cyber attackers. However, the proliferation of mobile devices such as smartphones, smartwatches, and tablets has made users of these devices a target for cyber attackers as well. However, the operating systems of personal computers and servers are generally different from those of mobile devices. This difference also leads to variations in attack types and malware types. In this study, a new ensemble method is proposed to detect malware in mobile devices with the Android operating system. The k-Nearest Neighbor (kNN), random forest, and model ensemble methods are combined with the Naive Bayes method. The proposed ensemble model has an accuracy of 98.69% and an F-score of 98.7%.