Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Henry T. Akaeze - Mississippi Valley State University
Co-Author(s): Serigne Sene, Dillard University, LA
Android smartphone applications today request permission to access a multitude of sensitive resources, which users must accept completely during installation. These permissions give the apps access to various sensors (e.g., microphone, GPS, and camera) and user data (e.g. contact, and photos). This, however, leaves these Android users vulnerable to nefarious apps that may be stealing from them. For enhanced privacy, we seek to leverage crowdsourcing to find a minimal set of permissions that will preserve the usability of the app for a diverse set of users. Therefore, we will give the participants of our study applications with different sets of permissions removed to get their feedback of how much these removed permissions affected the use of the app. We focused on Instagram which is a top photo/video sharing and social networking app. Using Android-apk tool and our lab-created script, we created different customized configurations of the Instagram app by removing one or more permissions from the manifest file, then recompiled and repacked the apps. We made a survey using Qualtrics, where a study will be run using Amazon Mechanical Turkers. In this study we plan to disperse various versions of different apps, starting with Instagram, to the survey participants. These versions will be the 15 configurations we get from removing four basic permissions (access to contacts, camera, location and microphone) from the apps using a lattice structure form. Finally, the users will be asked to take a survey summarizing their response on the app they just used. We will then from their feedback get the minimal sets of permissions various types of users will need in using the tested app. Due to time constraints, this survey has not been launched yet. A comprehensive result will be obtained once it is, and we will act on the obtained feedback from the Turkers.
Funder Acknowledgement(s): This work was supported by grants 1228364 and 1228471 from NSF. Thanks to Dr. Apu Kapadia and Qatrunnada Ismail for their guidance through this research process, and also to Robert Doyle, Michael K. Reiter and Srijita Das for their contributions. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the sponsors.
Faculty Advisor: Timothy Holston,