Danger theory-based privacy protection model for Social Networks

Abstract

Privacy protection issues in Social Networking Sites (SNS) usually raise from insufficient user privacy control mechanisms offered by service providers, unauthorized usage of user's data by SNS, and lack of appropriate privacy protection schemes for user's data at the SNS servers. In this paper, we propose a privacy protection model based on danger theory concept to provide automatic detection and blocking of sensitive user information revealed in social communications. By utilizing the dynamic adaptability feature of danger theory, we show how a privacy protection model for SNS users can be built with system effectiveness and reasonable computing cost. A prototype based on the proposed model is constructed and evaluated. Our experiment results show that the proposed model achieves 88.9% detection and blocking rate in average for user-sensitive data revealed by the services of SNS.