The number of Android applications (apps) is constantly increasing, by bringing new functionalities on users' smartphones and tablets every- day. Unfortunately, several apps pose many risks to the users, e.g., by including code that threaten user privacy or system integrity. Currently, most of the current security countermeasures for detecting dangerous apps show some weaknesses, mainly related to users' understanding and accep- tance. For these reasons, users would benet from an eective but simple technique that indicates whether an app is safe or risky to be installed. In this paper, we present MAETROID (Multi-criteria App Evaluator of TRust in AndrOID), a framework to evaluate the trustworthiness of An- droid apps, i.e. the amount of risk they pose to the users, e.g. in terms of condentiality and integrity. The framework performs a multi-criteria analysis of an app at deploy-time and returns a single easy-to-understand evaluation on the app's risk level, aimed at driving the user decision on whether installing or not a new app. The used criteria include the set of requested permissions and a further set of metadata retrieved from the marketplace, which denote the app quality and popularity. We have clas- sied 11,000 Android apps coming from Google Play and from a database of known malware. In particular, MAETROID has recognized as dan- gerous all the apps belonging to the database of malicious apps, while about 20% of apps from Google Play have been classied as medium risk. To evaluate the users' experience in the usage of MAETROID, we have collected and analyzed data from a survey that was completed by 189 sub- jects. The survey has measured the users' response to MAETROID, and has shown that MAETROID is more eective than the standard Android permission system in informing the user about an app risk level. Over the whole set of interviewees, MAETROID has been able to drive correctly the user in the decision of installing an app in more than 90% of cases.
Risk Analysis of Android Applications: A Multi-Criteria and Usable Approach
Fabio Del Bene;Fabio Martinelli;Ilaria Matteucci;Marinella Petrocchi;Andrea Saracino;
2015
Abstract
The number of Android applications (apps) is constantly increasing, by bringing new functionalities on users' smartphones and tablets every- day. Unfortunately, several apps pose many risks to the users, e.g., by including code that threaten user privacy or system integrity. Currently, most of the current security countermeasures for detecting dangerous apps show some weaknesses, mainly related to users' understanding and accep- tance. For these reasons, users would benet from an eective but simple technique that indicates whether an app is safe or risky to be installed. In this paper, we present MAETROID (Multi-criteria App Evaluator of TRust in AndrOID), a framework to evaluate the trustworthiness of An- droid apps, i.e. the amount of risk they pose to the users, e.g. in terms of condentiality and integrity. The framework performs a multi-criteria analysis of an app at deploy-time and returns a single easy-to-understand evaluation on the app's risk level, aimed at driving the user decision on whether installing or not a new app. The used criteria include the set of requested permissions and a further set of metadata retrieved from the marketplace, which denote the app quality and popularity. We have clas- sied 11,000 Android apps coming from Google Play and from a database of known malware. In particular, MAETROID has recognized as dan- gerous all the apps belonging to the database of malicious apps, while about 20% of apps from Google Play have been classied as medium risk. To evaluate the users' experience in the usage of MAETROID, we have collected and analyzed data from a survey that was completed by 189 sub- jects. The survey has measured the users' response to MAETROID, and has shown that MAETROID is more eective than the standard Android permission system in informing the user about an app risk level. Over the whole set of interviewees, MAETROID has been able to drive correctly the user in the decision of installing an app in more than 90% of cases.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
