The aim of this competition is to improve the efficiency and effectiveness of Innovate UK’s operational functions.
We welcome as many ideas as possible for the application of machine learning to Innovate UK’s existing data. You should show how you would develop and test a prototype application to:
- improve operational efficiency
- improve the efficiency and quality of decision-making, both pre and post-funding award
We have identified 2 areas where machine learning could help us reduce costs and work more efficiently:
1. Assessor allocation: identifying key words from funding applications to allocate appropriate assessors. Currently we use a time-consuming manual allocation process. Using machine learning would help us move towards more automated assessor allocation. It would also lead to improved quality and robustness in the assessment process. We would save money and allow our teams to focus on more value-added activities. This would help us fund successful projects more quickly.
2. Checking for undeclared re-submissions: searching for re-submissions, duplicate and reassessed applications. Currently we do this manually. We could save money by automating this process and reducing the possibility of fraud.
These are 2 possibilities and there are more below but we welcome other ideas:
- analysing competition queries, complaints and enquiries data to improve the quality of our communications
- gathering information automatically from different styles of application form and representing that data in a meaningful and readily-accessible way
- learning from project outcomes to better inform competition and project design
- learning from economic exploitation data to better inform investment decisions
- using monitoring reporting data to inform continuous improvement of monitoring services
- reducing the potential for fraud
- improving the accuracy of expenditure claims forecasting helping us to schedule projects more effectively
Our aim is for efficiency savings to cover the cost of the project investment.
This is phase 1 of a potential 2-phase competition. A decision to proceed with phase 2 will depend on the outcomes from phase 1. Only successful applicants from phase 1 will be able to take part in phase 2. We may change the budget and transfer funding between the different phases.
We will provide details of available data to help you develop your proposal. We will provide successful applicants with anonymised data for their feasibility study. We will also provide descriptions of our internal systems. Successful applicants will need to sign a non-disclosure agreement.
In phase 2 the successful applicants will continue to develop and test their prototype.