Analyzing Effectiveness of Traditional Privacy Preserved Data Anonymization Techniques

Abstract

The rapid advances in technology and the increase in cybercrimes and data mishandling, raises substantial apprehensions about data privacy and security for users globally. The widespan use of social networking sites and many other online services has generated vast amounts of data, contributing to big data. Accordingly, there is a increasing demand for PPDP - Privacy Preserving Data Publishing and it has gathered significant attention due to the sensitivity of identifiable data and the need to balance privacy concerns with the effective use of the publicly available data of users. To defend the privacy of data owners, various anonymization techniques are applied before data is shared to a third party or used for analytics. In this paper, we are analyzing some of the traditional anonymization techniques based on utility metrics and study how well these methods can prevent reidentification attacks.