A Machine Learning Approach to Android Malware Detection
Authors:
- Justin SahsJSJustin Sahs
- Latifur KhanLKLatifur Khan
- 1 August 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 141-147
Abstract
With the recent emergence of mobile platforms capable of executing increasingly complex software and the rising ubiquity of using mobile platforms in sensitive applications such as banking, there is a rising danger associated with malware targeted at mobile devices. The problem of detecting such malware presents unique challenges due to the limited resources avalible and limited privileges granted to the user, but also presents unique opportunity in the required metadata attached to each application. In this article, we present a machine learning-based system for the detection of malware on Android devices. Our system extracts a number of features and trains a One-Class Support Vector Machine in an offline (off-device) manner, in order to leverage the higher computing power of a server or cluster of servers.Keywords
This publication has 19 references indexed in Scilit:
- Location based power analysis to detect malicious code in smartphones2011
- Crowdroid2011
- LIBSVMACM Transactions on Intelligent Systems and Technology, 2011
- VirusMeter: Preventing Your Cellphone from Spies2009
- Architecture of a morphological malware detectorJournal of Computer Virology and Hacking Techniques, 2008
- A Neyman–Pearson Approach to Statistical LearningIEEE Transactions on Information Theory, 2005
- Data mining methods for detection of new malicious executables2002
- Neural networks for computer virus recognitionIEEE Expert, 1996
- MCF: a malicious code filterComputers & Security, 1995
- XVI. Functions of positive and negative type, and their connection the theory of integral equationsPhilosophical Transactions of the Royal Society A, 1909