This talk presents the issue of extracting insights from data sets containing sensitive information, while preserving privacy. We will review the techniques used today for private data sharing and their limitations. We will explain why it is hard to share useful data while preserving the privacy and point to promising approaches enabling the implementation of such a goal.
In this talk we will present the often overlooked issue of privacy preservation while extracting or sharing insights from data sets containing sensitive information. We will review the various techniques used today for privacy-preserving data sharing (from k-anonymity to differential privacy in their different flavours), and discuss the strengths and weaknesses of each. We will explain why it is hard to share useful data in all cases while preserving the privacy and what can be done to mitigate this. We will finish by pointing to a selection of promising approaches enabling the implementation of privacy-preserving data sharing.