Permission-based Risk Signals for App Behaviour Characterization in Android Apps

Olukoya, O. , Mackenzie, L. and Omoronyia, I. (2018) Permission-based Risk Signals for App Behaviour Characterization in Android Apps. In: 5th International Conference on Information Systems Security and Privacy, Prague, Czech Republic, 23-25 Feb 2019, pp. 183-192. ISBN 9789897583599 (doi: 10.5220/0007248701830192)

Abstract

With the parallel growth of the Android operating system and mobile malware, one of the ways to stay protected from mobile malware is by observing the permissions requested. However, without careful consideration of these permissions, users run the risk of an installed app being malware, without any warning that might characterize its nature. We propose a permission-based risk signal using a taxonomy of sensitive permissions. Firstly, we analyse the risk of an app based on the permissions it requests, using a permission sensitivity index computed from a risky permission set. Secondly, we evaluate permission mismatch by checking what an app requires against what it requests. Thirdly, we evaluate security rules based on our metrics to evaluate corresponding risks. We evaluate these factors using datasets of benign and malicious apps (43580 apps) and our result demonstrates that the proposed framework can be used to improve risk signalling of Android apps with a 95% accuracy.

• Item Type: Conference Proceedings
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Omoronyia, Dr Inah and Mackenzie, Dr Lewis and Olukoya, Mr Oluwafemi
• Authors: Olukoya, O., Mackenzie, L., and Omoronyia, I.
• College/School: College of Science and Engineering > School of Computing Science
• ISBN: 9789897583599
• Copyright Holders: Copyright © 2018 ICISSP
• First Published: First published in Proceedings of the 5th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP
• Publisher Policy: Reproduced with the permission of the publisher

Related URLs

• Organisation