Permission based detection system for android malware


Creative Commons License

Utku A., Dogru İ. A.

JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY, cilt.32, sa.4, ss.1015-1024, 2017 (SCI-Expanded) identifier identifier

Özet

Nowadays uses and functionality of mobile devices show a large increase. Mobile devices have become platforms where users keep personal information. Due to such characteristics, mobile devices have become the target of the attackers. In this study, a permission-based malware detection system have been developed using Naive Bayes and KNN algorithms. The test results of the proposed system were analyzed and compared for each algorithm. Analysis results showed that Naive Bayes classifier was successful in detecting malware with 97.29% And Knn classifier was successful in detecting malware with 97.74%.