Native malware detection in smartphones with android OS using static analysis, feature selection and ensemble classifiers
Authors:
- S. Morales-OrtegaSMS. Morales-Ortega
- P.J. Escamilla-AmbrosioPEP.J. Escamilla-Ambrosio
- A. Rodriguez-MotaARA. Rodriguez-Mota
- L.D. Coronado-De-AlbaLCL.D. Coronado-De-Alba
- 1 October 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
The use of Smartphones (SPs)with Android Operating System (AOS) has reached unprecedented popularity. This is due to the many features that these devices offer as Internet connection, storage of information as well as the ability to perform diverse online transactions. As a result, these devices have become the main target of malware attacks that try to exploit the security vulnerabilities of AOS.Therefore, in order to mitigate these attacks, methods for malware analysis and detection are needed.In this work a method for analysis and detection of malware, which can run natively in the device, is proposed. The approach can analyze applications already installed on the device, monitor new apps installations or updates. Static analysis is used to determine the permissions, hardware and software features requested by applications. An application being analyzed is classified as malware or benign using a model based on ensemble machine learning classifiers and feature selection algorithms. To validate the proposed method, 1377 malware samples and 1377 benign samples, collected from different sources, were used.Results show that the proposed approach detects malware with 96.26%of accuracy. Additional tests were conducted in three different SPs devices to validate malware detection performance in a real environment andto obtain an average execution time. Results of these tests show that the proposed method detects malware with 94.48% of accuracy, getting the analysis results of an application in 35milliseconds.Keywords
This publication has 10 references indexed in Scilit:
- A review on feature selection in mobile malware detectionDigital Investigation, 2015
- APK Auditor: Permission-based Android malware detection systemDigital Investigation, 2015
- A static Android malicious code detection method based on multi-source fusionSecurity and Communication Networks, 2015
- Android Security: A Survey of Issues, Malware Penetration, and DefensesIEEE Communications Surveys & Tutorials, 2014
- Drebin: Effective and Explainable Detection of Android Malware in Your Pocket2014
- Dissecting Android Malware: Characterization and Evolution2012
- “Andromaly”: a behavioral malware detection framework for android devicesJournal of Intelligent Information Systems, 2011
- Ensemble-based classifiersArtificial Intelligence Review, 2009
- Random ForestsMachine Learning, 2001
- Improvements to Platt's SMO Algorithm for SVM Classifier DesignNeural Computation, 2001