search

alan-turing-institute / QUIPP-collab

Collaboration on the QUIPP project
1 stars 1 forks source link

Accenture Data Sharing call for proposals #160

Closed martintoreilly closed 3 years ago

martintoreilly commented 3 years ago

This Accenture call looks like a good fit for getting some follow on funding.

Accenture Turing Strategic Partnership: Call for Expressions of Interest - Data Sharing

The Turing recently launched a five-year Strategic Partnership with Accenture to advance data science and AI research with a focus on delivering substantial business and societal value. Through the partnership we are seeking to fund projects that fall under the research theme ‘Data Sharing’ (incorporating Synthetic Data, PETs and FEDS). We have £300,000 to award through this call. In special circumstances there may be an opportunity to increase the funding awarded for an outstanding project.

Expressions of Interest should be completed and submitted by Thursday 22 April at midday BST. Please read the full guidance for further details, including examples of possible research directions, and assessment and eligibility criteria. Visit the website to find out more and apply.

martintoreilly commented 3 years ago

Looks like a good fit for supporting an extension of QUIPP to work on the CC-HIC data with UCL.

martintoreilly commented 3 years ago

We've a draft letter to UCL describing our proposal for accessing and using the CC-HIC.

martintoreilly commented 3 years ago

@gmingas notes that the Accenture call is probably not a good fit for the UCL CC-HIC health dataset. What are people's thoughts on what might work for us with a Finance, Economics or Social Sciences focus?

I just read the full Accenture call guidance, it seems the funding is supposed to be given to projects with a Finance, Economics or Social Sciences element (aligned to the Accenture Turing Strategic Partnership). So I think the UCL collaboration is out of scope for this one.

martintoreilly commented 3 years ago

Re-reading the call guidance closely, I think we could still make a case for the value in validating synthetic data generation methods and measures of privacy/utility on a real world data set, even if it is not a finance, economics or social science dataset. The references to the domain focus are few and in the overview of the programme. The detail on the Accenture partnership and call focus doesn't mention domain at all.