In this paper, we propose a randomization scheme, LORA (Link Obfuscation by Randomization), to obfuscate edge existence in graphs. Specifically, we extract the source graph's hierarchical random graph model and reconstruct the released graph randomly with this model. We show that the released graph can preserve critical graph statistical properties even after a large number of edges have been replaced. To measure the effectiveness of our scheme, we introduce the notion of link entropy to quantify its privacy-preserving strength wrt the existence of edges. © 2011 Springer-Verlag.
CITATION STYLE
Xiao, Q., Wang, Z., & Tan, K. L. (2011). LORA: Link obfuscation by randomization in graphs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6933 LNCS, pp. 33–51). Springer Verlag. https://doi.org/10.1007/978-3-642-23556-6_3
Mendeley helps you to discover research relevant for your work.