Poster : Optimization based Data De-anonymization

In this poster, we study optimization based structural data De-Anonymization (DA), including social data, mobility traces, etc. We make a DA practice by presenting a novel single-phase cold start Optimization based DA (ODA) algorithm followed by theoretical and experimental analysis. Experimental resutls of ODA show that about 77.7%− 83.3% of the users in Gowalla (0.2M users and 1M edges) [5] are de-anonymizable, which implies optimization based DA is implementable and powerful in practice. Furthermore, We discuss the future research directions of this project.