Differential Privacy on Edge Computing

This paper presents an overview of privacy protection, with a focus on differential privacy (DP), from the perspective of edge computing. It explores the application of DP in various associative analysis techniques, including heavy hitter mining, frequent itemset mining, and association rules mining, within the context of edge computing. The paper also highlights the current challenges and future research directions in this area, including differentially private hybrid models and federated learning. By examining the intersection of privacy protection and edge computing, this paper provides insights into the application of DP and its potential for preserving privacy in associative analysis tasks within edge computing environments.