On the efficacy of static features to detect malicious applications in Android

The Android OS environment is today increasingly targeted by malware. Traditional signature based detection algorithms are not able to provide complete protection especially against ad-hoc created malwares. In this paper, we introduce an anomaly-based approach to assess whether an application is malicious or not on the basis of applications permission and related APIs exploiting the advantages of machine learning algorithms combining different fusion rules. We study the performance of our approach in terms of false alarms tradeoff. Results demonstrate that our approach reach an equal error rate as little as 5.59% depending on the configuration.

GENEIATAKIS Dimitrios;  SATTA Riccardo;  NAI FOVINO Igor;  NEISSE Ricardo; 

2016-01-21

Springer International Publishing

JRC95053

978-3-319-22905-8,   

0302-9743,   

http://link.springer.com/chapter/10.1007%2F978-3-319-22906-5_7,    https://publications.jrc.ec.europa.eu/repository/handle/JRC95053,   

10.1007/978-3-319-22906-5_7,