A Systematic Review on Privacy Preserving Data Publishing Techniques

What is Privacy? -According to the definition from Cambridge dictionary, "Someone's right to keep their personal matters and relationships secret". The term privacy is defined as an action where the data is kept hidden from either anonymous user, server to avoid use of malpractice of the data. Healthcare data is considered as a most significant but sensitive data in the world, since it has all private information about patient such as diseases, treatment, prescription, name, address etc. The volume of the data generated in healthcare industry is rapidly growing. In this patient centric world, to get effective results, we need to increase healthcare data utility. With increasing data utility , the privacy of the same data is compromised which is the another important challenge that users and healthcare data publishers are facing, Since there is no monitory control on data which is published on internet. Hiding sensitive healthcare data from either untrusted users or third party data publishers is an important concern today. Healthcare Data Publishing is the process where certain transformation (such as anonymization, generalization, suppression etc.) can be applied before publishing healthcare data online. From the available research, it is seen that such transformations are not susceptible to certain attacks like background knowledge, homogeneity etc. This review paper studies all existing Privacy Preserving Data Publishing (PPDP) schemes using data generalization. The literature review also touches recent researches on ARX tool -which is an open source data de-identification tool for analysing risk and utility factor of healthcare data. The paper finally concludes with feasible research gaps from available literature survey.