Research and application of privacy protection technology based on big data environment

Abstract

With the development of the Internet of Things, the popularization of the mobile Internet, and the rapid promotion of social networks, the growth of data has entered a big explosion, and the development of information technology has caused a torrent of big data. The continuous development of technology brings people convenience, speed and comfort, but also hides hidden worries. Combined with differential privacy and clustering, a clustering-based differential privacy universal dataset release method for mixed datasets is proposed: using k-prototype clustering algorithm to group records in the hybrid dataset to reduce differential privacy Query sensitivity and the amount of noise to be added to improve data utility while providing data privacy protection; perform attribute difference calculations for numerical attributes and categorical attributes combined with weights, and measure their information loss separately. Finally, Experimental results show that this algorithm can improve the usability of data publishing.