Graph-based Clustering under Differential Privacy
Data Structures and Algorithms
2018-03-13 v1 Machine Learning
Abstract
In this paper, we present the first differentially private clustering method for arbitrary-shaped node clusters in a graph. This algorithm takes as input only an approximate Minimum Spanning Tree (MST) released under weight differential privacy constraints from the graph. Then, the underlying nonconvex clustering partition is successfully recovered from cutting optimal cuts on . As opposed to existing methods, our algorithm is theoretically well-motivated. Experiments support our theoretical findings.
Cite
@article{arxiv.1803.03831,
title = {Graph-based Clustering under Differential Privacy},
author = {Rafael Pinot and Anne Morvan and Florian Yger and Cédric Gouy-Pailler and Jamal Atif},
journal= {arXiv preprint arXiv:1803.03831},
year = {2018}
}