For many years, the field of privacy-preserving data mining (PPDM) focused mostly on the perturbation and
anonymization of tabular data. However, nowadays with the rapid growth of social networks and online
communities such as Facebook, MySpace, and LinkedIn a new area of research in PPDM has attracted
the attention of community of privacy researchers: Privacy preserving analysis of social networks data which
are in the form of large and complex graphs. In this seminar we will talk about different aspects of privacy threats
in social networks as well as the challenges in anonymizing such network data. We will survey recently
developed techniques in this area and we will present our new approach for privacy preserving association
rule mining in social networks.