Privacy Preserving Publishing of Social Network Data Privacy and Big Data Mining

Privacy is a major concern in big data mining. Advances in development of data mining technologies bring serious risk to the security of individual’s sensitive data. An emerging research in data mining, known as privacypreserving data mining (PPDM), has been broadly studied in recent years. The basic idea of PPDM is to transform or change the data in such a way so as to not to compromise on security of individual’s sensitive data and also to perform data mining algorithms effectively. In this paper, I have identified four different category of users who are involved in data mining applications and privacy concerns of each category. Basically, any data mining application will have four kind of users namely, data provider, data collector, data miner and decision maker. I briefly introduce the basics of related research topics and current approaches, and present some basic thoughts on future research. By differentiating the responsibilities of different users with respect to security of sensitive information, I would like to provide some useful insights into the study of PPDM on social networking data. One main characteristic of social networks is that they keep evolving over time. The data collector needs to publish the network data periodically. The privacy issue in sequential publishing of dynamic social network data has recently attracted researchers' attention. Keywords— Data mining, sensitive data, privacypreserving data mining, and social network