Research on the Application of Local Differential Privacy: A Survey

In this age of information explosion, people are fully aware of the great value of data and pay more and more attention to protecting personal privacy. Therefore, fully exploring the value of data without revealing user privacy is an urgent problem that needs to be solved in big data. The differential privacy protection technology proposed in 2006 is widely used in the security processing of massive data because of its strict mathematical definition. However, differential privacy technology is a privacy-protecting mechanism based on trusted third parties. Local differential privacy technology was proposed in 2008 to reduce the independence of trusted third parties. Since its introduction, local differential privacy technology has been widely used to solve privacy protection problems in various big data applications. In order to flexibly use local differential privacy technology to deal with the privacy protection problems in various new applications, this article elaborates in detail the existing research work of local differential privacy technology in the fields of data release, data mining, and machine learning, and summarizes the local The application framework of differential privacy in these fields and their respective advantages and disadvantages finally address the research challenges that local differential privacy technology still faces.