De-anonymization of heterogeneous random graphs in quasilinear time

Abstract

There are hundreds of online social networks with billions of users in total. Many such networks publicly release structural information, with all personal information removed. Empirical studies have shown, however, that this provides a false sense of privacy - it is possible to identify almost all users that appear in two such anonymized network as long as a few initial mappings are known. We analyze this problem theoretically by reconciling two versions of an artificial power-law network arising from independent subsampling of vertices and edges. We present a new algorithm that identifies most vertices and makes no wrong identifications with high probability. The number of vertices matched is shown to be asymptotically optimal. For an n-vertex graph, our algorithm uses n ε seed nodes (for an arbitrarily small ε) and runs in quasilinear time. This improves previous theoretical results which need Θ(n) seed nodes and have runtimes of order n 1+Ω(1). Additionally, the applicability of our algorithm is studied experimentally on different networks. © 2014 Springer-Verlag Berlin Heidelberg.