Smart Data Foundry data science team wins international synthetic data challenge
At Smart Data Foundry we're passionate about maintaining people’s privacy. Creating synthetic datasets helps to achieve this, by having the same essential patterns and relationships as the corresponding real datasets but containing only artificial people and events.
Working with synthetic data allows innovators and researchers to work safely with data without exposing sensitive information.
At Smart Data Foundry we test a range of methodologies, in order to build the most effective, useful, yet protected, datasets to work with. The ideal synthetic dataset looks just like a different sample from the same underlying population from which it has been generated.
This is new territory and Smart Data Foundry is rapidly gaining strength and depth in synthetic data.
Members of our Data Science team were recently invited to enter an international synthetic data challenge under our previous name of the Global Open Finance Centre of Excellence (GOFCoE) organised by the United Nations Economic Commission of Europe (UNECE) High Level Group for Modernisation of Official Statistics (HLG-MOS) in order to evaluate their upcoming Synthetic Data for National Statistical Offices: A Starter Guide, to be published later this year.
This was a great experience for the team, since rebranded as Smart Data Foundry, who pitched their expertise against a strong global field to explore and learn techniques for generating synthetic data.
Paola Arce, Victor Alfonzo Diaz and Euan Gardner from Smart Data Foundry were delighted to win the competition and took great encouragement that they’re gaining knowledge and experience in synthetic data through the process of the week-long challenge.
Smart Data Foundry is part of the University of Edinburgh and a collaboration with the Financial Data and Technology Association (FDATA) and FinTech Scotland. With roots in Scotland and supported by UK Research and Innovation’s flagship Strength in Places Fund, Smart Data Foundry has a close working relationship with the Scottish Government as it seeks to unlock the power of data as a force to improve people’s lives.
It’s fantastic to see talent from Scotland leading the way when it comes to synthetic data. This area is still relatively new, and I’m excited about the possibilities it offers for research and innovation, allowing for the quick and safe deployment of representative data that can be used for so many positive purposes.
Nick Radcliffe, Smart Data Foundry’s Chief Data Scientist added “I’m delighted to have come out on top in this competition against such strong teams and I’m proud of the team and how much they’re learning in this space. The competition also gave us the opportunity to test the GEMINAI software from Diveplane, which performed really well in the tests.”
Alan Cross, Diveplane Chief Commercial Officer stated “Smart Data Foundry used our award winning GEMINAI synthetic engine and I’m pleased that it performed so well and supported them to a successful outcome. We share common values, specifically, that data must be protected and used responsibly and to that end we are confident this partnership will succeed.”
Organised by the United Nations (UNECE) High Level Group for Modernisation of Official Statistics (HLG-MOS), Kate Burnett-Isaacs (Statistics Canada), added “This was a great week-long competition and congratulations to all that entered to understand and learn techniques for generating synthetic data. We’re working on compiling the information from everyone’s challenge contributions to create a readable summary of technical resources that will be available to all.”
All of the outcomes and evaluations from the challenge are now being used to update the HLG-MOS Synthetic Data For National Statistical Organizations: A Starter Guide, which will be published by the UNECE and publicly available later this year.