Authors:
Rosangela Casolare
1
;
Giacomo Iadarola
2
;
Fabio Martinelli
2
;
Francesco Mercaldo
3
;
2
and
Antonella Santone
3
Affiliations:
1
Department of Biosciences and Territory, University of Molise, Pesche (IS), Italy
;
2
Institute of Informatics and Telematics, National Research Council of Italy, Pisa, Italy
;
3
Department of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, Campobasso, Italy
Keyword(s):
Malware, Audio, Android, Machine Learning, Deep Learning, Security, Classification.
Abstract:
Nowadays smartphones, and generically speaking mobile devices, allow users a plethora of tasks in total mobility for instance, from checking the balance on the bank account to distance learning. In this context it is really critical the detection of malicious behaviours, considering the weaknesses of the current antimalware mechanisms. In this paper we propose a method for malicious family detection exploiting audio signal processing: in fact, an application is converted into an audio file and then is processed to generate a feature vector to input several classifiers. We perform a real-world experimental analysis by considering a set of malware targeting the Android platform i.e., 4746 malware belonging to 10 families, showing the effectiveness of the proposed approach for Android malicious family detection.