학술논문

A Profile-Based Privacy Protection Method using Sandbox Environment and k-Anonymity: Computer Data Privacy
Document Type
Conference
Source
2023 International Conference on Communication, Security and Artificial Intelligence (ICCSAI) Communication, Security and Artificial Intelligence (ICCSAI), 2023 International Conference on. :119-123 Nov, 2023
Subject
Communication, Networking and Broadcast Technologies
Computing and Processing
General Topics for Engineers
Robotics and Control Systems
Signal Processing and Analysis
Privacy
Data privacy
Machine learning algorithms
Information sharing
Machine learning
Information leakage
Smart phones
Android
Privacy protection
Sandboxing
Profile management
Data masking
k-anonymity
Language
Abstract
Android apps request excessive permissions to get the access of sensitive information, some apps request excessive permissions which creates a seriously threat to the user privacy and security. This research is based on profile-based privacy protection method for Android apps which integrate principle of sandboxing and data masking techniques. User can manage multiple profiles for different purpose as per requirement. Each profile has assigned a set of rules, permissions and data. They don't share data with other profiles. Sandbox environment help in keeping the each profile isolated from each other. To protect the data of each user, masking of data has been done using $k$ anonymity. The proposed method has been assessed with different app categories and for different user profiles. The results shows that the method can effectively protect users' privacy and data, while introducing acceptable performance overhead.