Abstract
This disclosure describes the use of clustering techniques to group similar apps on a platform and identify anomalous permissions. Using the described techniques, platform providers can ensure that apps only obtain access to user data that is essential to provide their functionality. Further, platform providers can modularize their API, e.g., break down permissions or features that apps can access, to limit app access to user data. Further, on platforms that conduct human review of apps, app clustering results can be used to provide reviewers with information regarding how the data accesses by a particular app compare with other similar apps. The clustering results can also be utilized to provide app developers suggestions for use of the platform API and data access permissions.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Anonymous, "App Clustering to Identify Anomalous Data Access Permissions", Technical Disclosure Commons, (April 20, 2020)
https://www.tdcommons.org/dpubs_series/3159