An incentive-based protection and recovery strategy for secure big data in social networks

Abstract

Big data sources, such as smart vehicles, IoT devices, and sensor networks, differ from traditional data sources in both output volume and variety. Big data is usually stored on various types of network nodes, which is prone to data security and privacy problems, such as virus infection. In particular, the spread of viruses through social networks can cause large-scale destruction and privacy leakage in the network. This paper aims to provide a solution to protect the security of big data. First, the users are divided into five states according to their reactions to data virus: susceptible, contagious, doubt, immune, and recoverable. Then, we propose a novel model for studying the propagation mechanism of data virus. To control the spread of virus and protect data security, an incentive mechanism is introduced. After that, a protection and recovery strategy (PRS) is developed to reduce infected users and increase the immunized. The experimental results on two data sets indicate that our model gives a good description of the data virus propagation process, and PRS is better than both acquaintance immunization and target immunization methods, which validates the privacy preserving strategy for big data in networks.