Anonymizing Weighted Social Network Graphs

TitleAnonymizing Weighted Social Network Graphs
Publication TypeConference Paper
Year of Publication2010
AuthorsDas, S, Eğecioğlu Ö, El Abbadi A
Conference NameICDE
Pagination904 - 907
Date Published03/2009
Place PublishedLong Beach, LA

The increasing popularity of social networks has initiated a fertile research area in information extraction and data mining. Although such analysis can facilitate better understanding of sociological, behavioral, and other interesting phenomena, there is growing concern about personal privacy being breached, thereby requiring effective anonymization techniques. In this paper, we consider edge weight anonymization in social graphs. Our approach builds a linear programming (LP) model which preserves properties of the graph that are expressible as linear functions of the edge weights. Such properties form the foundations of many important graph-theoretic algorithms such as shortest paths, k-nearest neighbors, minimum spanning tree, etc. Off-the-shelf LP solvers can then be used to find solutions to the resulting model where the computed solution constitutes the weights in the anonymized graph. As a proof of concept, we choose the shortest paths problem, and experimentally evaluate the proposed techniques using real social network data sets.

Refereed DesignationRefereed