Explaining Unwanted Behaviours in Context

Wei Chen; David Aspinall; Andrew Gordon; Charles Sutton; Igor Muttik

Explaining Unwanted Behaviours in Context

Wei Chen
, David Aspinall
, Andrew Gordon
, Charles Sutton
, Igor Muttik

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Mobile malware has been increasingly identified based on unwanted behaviours like sending premium SMS messages. However, unwanted behaviours for a group of apps can be normal for another, i.e., they are context sensitive. We develop an approach to automatically explain unwanted behaviours in context and evaluate the automatic explanations via a user-study with favourable results. These explanations not only state whether an app is malware but also elaborate how and in what kind of context a decision was made.

Original language	English
Title of host publication	Proceedings of the 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems (ESSoS 2016)
Publisher	CEUR-WS.org
Pages	38-45
Number of pages	8
Publication status	Published - 6 Apr 2016
Event	1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems - London, United Kingdom Duration: 6 Apr 2016 → 6 Apr 2016 http://ceur-ws.org/Vol-1575/

Conference

Conference	1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems
Abbreviated title	IMPS 2016
Country/Territory	United Kingdom
City	London
Period	6/04/16 → 6/04/16
Internet address	http://ceur-ws.org/Vol-1575/

Access to Document

IMPS_2016_paper_1-2Final published version, 163 KB

http://ceur-ws.org/Vol-1575/

App Guarden: Resilient Application Stores
Aspinall, D. (Principal Investigator), Franke, B. (Co-investigator), Gordon, A. (Co-investigator), Sannella, D. (Co-investigator), Stark, I. (Co-investigator) & Sutton, C. (Co-investigator)
EPSRC
1/09/13 → 31/08/16
Project: Research

Cite this

@inproceedings{781cfc7e1cba4948ab9a38b730cff09d,

title = "Explaining Unwanted Behaviours in Context",

abstract = "Mobile malware has been increasingly identified based on unwanted behaviours like sending premium SMS messages. However, unwanted behaviours for a group of apps can be normal for another, i.e., they are context sensitive. We develop an approach to automatically explain unwanted behaviours in context and evaluate the automatic explanations via a user-study with favourable results. These explanations not only state whether an app is malware but also elaborate how and in what kind of context a decision was made.",

author = "Wei Chen and David Aspinall and Andrew Gordon and Charles Sutton and Igor Muttik",

year = "2016",

month = apr,

day = "6",

language = "English",

pages = "38--45",

booktitle = "Proceedings of the 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems (ESSoS 2016)",

publisher = "CEUR-WS.org",

note = "1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems, IMPS 2016 ; Conference date: 06-04-2016 Through 06-04-2016",

url = "http://ceur-ws.org/Vol-1575/",

}

Chen, W, Aspinall, D, Gordon, A, Sutton, C & Muttik, I 2016, Explaining Unwanted Behaviours in Context. in Proceedings of the 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems (ESSoS 2016). CEUR-WS.org, pp. 38-45, 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems, London, United Kingdom, 6/04/16. <http://ceur-ws.org/Vol-1575/>

Explaining Unwanted Behaviours in Context. / Chen, Wei; Aspinall, David; Gordon, Andrew et al.
Proceedings of the 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems (ESSoS 2016). CEUR-WS.org, 2016. p. 38-45.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Explaining Unwanted Behaviours in Context

AU - Chen, Wei

AU - Aspinall, David

AU - Gordon, Andrew

AU - Sutton, Charles

AU - Muttik, Igor

PY - 2016/4/6

Y1 - 2016/4/6

N2 - Mobile malware has been increasingly identified based on unwanted behaviours like sending premium SMS messages. However, unwanted behaviours for a group of apps can be normal for another, i.e., they are context sensitive. We develop an approach to automatically explain unwanted behaviours in context and evaluate the automatic explanations via a user-study with favourable results. These explanations not only state whether an app is malware but also elaborate how and in what kind of context a decision was made.

AB - Mobile malware has been increasingly identified based on unwanted behaviours like sending premium SMS messages. However, unwanted behaviours for a group of apps can be normal for another, i.e., they are context sensitive. We develop an approach to automatically explain unwanted behaviours in context and evaluate the automatic explanations via a user-study with favourable results. These explanations not only state whether an app is malware but also elaborate how and in what kind of context a decision was made.

M3 - Conference contribution

SP - 38

EP - 45

BT - Proceedings of the 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems (ESSoS 2016)

PB - CEUR-WS.org

T2 - 1st International Workshop on Innovations in Mobile Privacy and Security co-located with the International Symposium on Engineering Secure Software and Systems

Y2 - 6 April 2016 through 6 April 2016

ER -

Explaining Unwanted Behaviours in Context

Abstract

Conference

Access to Document

Fingerprint

Projects

App Guarden: Resilient Application Stores

Cite this