Sensitive Knowledge Hiding Application

Abstract

Diverse organizations store their business related data in their own databases. Some of them publish their data as is to third parties for research purposes. However, this way of data sharing has the risk of disclosing sensitive knowledge which the database may entail. For safe publishing the database must be sanitized before sharing; so that no sensitive knowledge is conveyed. The process is known as privacy-aware data publishing in the literature. The process requires (i) identification of sensitive knowledge, and (ii) removal of them from the to-be-released database. The challenge here is to hide all sensitive knowledge, and at the same time keep the released version as similar as possible to the original one. In the literature, there exists separate proposals to hide various kinds of sensitive knowledge formats from diverse databases. In this work, we present a single unified platform for this task.