Local Differential Privacy Preserving in Social Networks
Primary Investigator:
Feng Li
Tianchong Gao; Feng Li
Abstract
Online social networks often contain sensitive information about individuals. Therefore, de-anonymizing the graph before releasing becomes an important issue. We define the notion of group based local differential privacy. In particular, by resolving the network into 1-neighborhood graphs and applying HRG-based methods, local differential privacy reduces the noise scale. By grouping the HRGs’ output spaces with overlap together and sampling an outstanding HRG, we perturbed the subgraphs in a group to be extremely similar to enhance the privacy power beyond differential privacy..