GB-London: Tagging Project using Artificial Intelligence

Over our 30-year history, we have built a rich mine of grant data from the heritage projects we have funded. Whilst we have a number of closed-category fields and existing taxonomies we use for regular reporting, historically, we have had to manually tag and categorise this data to support broader reporting requirements. This is a time-consuming and resource-intensive process. To aid this process, we want to test and pilot the possibility of using artificial intelligence (AI) tools to categorise historic (and evaluate its adaptability to incoming) grant data to meet our current and future thematic reporting requirements. We think this can be achieved through the application of targeted 'tags' applied to project information that is collected via both free text and closed-response fields. We are keen to test this by examining how AI could help across two specific sub-sets of our grant data, namely (i) grants relating to places of worship and (ii) information about the intended and actual beneficiaries of our grants. This tender is split into two phases with a possible break clause after the discovery element. This will allow us to assess our requirements and develop a specification to guide us on the viability of proceeding to a possible proof-of-concept phase. In order to decide whether to move forward to a proof-of-concept and development of a trial product, the contractor will be required to use learnings from the discovery phase to demonstrate how the solution would work in practice with our workflows and activities using the subsets of data mentioned above (places of worship and beneficiary data) as an example.