Recent years have seen the idea of Big Data come to prominence: the collection and analysis of very large data sets, used in a wide range of application areas. Many such data sets, in their raw form, contain personal or sensitive information about individuals, raising privacy concerns. These concerns have given rise to new topics in statistical methodology. which aim to combat this issue. This workshop takes the opportunity to take a wide view on all aspects of "privacy and security in statistics". At present, differential privacy is probably the largest area to tackle these questions and is quite established, but there are other complementary relevant emerging areas.

This full day workshop, organised jointly between the Emerging Applications Section of the Royal Statistical Society and the Statistics section, Imperial College London, will cover the following areas.

  • Differential privacy for preserving data privacy in the outputs of analysis
  • Issues around anonymity and anonymisation of raw data
  • Cryptographic security using recent encryption schemes with homomorphic properties
  • Building statistical models encrypted
We welcome attendees from academia, industry and governmental organisations.

This event is co-organised by:

Additionally, we are very grateful for sponsorship from the Data Science Institute, Imperial College London.