Protecting User Privacy in Mobile Environment Using ELM-UPP

Abstract

In this paper, we address the topic of user privacy preservation in mobile environment. Existing techniques mostly rely on structure-based spatial cloaking, but pay little attention to location semantic information. Yet, such information may disclose sensitive information about mobile users. Thus, we propose ELM-UPP, a semantic-awareness privacy preservation framework to protect users privacy from violation. It allows mobile users to explicitly define their preferred privacy requirements in terms of location hiding measures. To provide location semantic protection for mobile users, in our framework, two features are firstly proposed to capture the semantic of locations. Then, a ELM-based unsupervised clustering is leveraged to detect semantic homogeneity locations. Besides, we design cloaking areas that should cover different semantic locations as well as achieving high quality of service. Since the problem of calculating the optimal cloaking area is NP-hard, we design a greedy algorithm that balances quality of service and privacy requirements. Extensive experimental results show the efficiency and effectiveness of our proposed algorithm.