Ventures

What are you looking for?

University of Nottingham > INFINITY > Ventures

Find out more about the projects the University of Nottingham Inclusive Financial Technology Hub has been involved in.

DECSYS

Key expertise from UoN

Explainable Artificial Intelligence, Uncertainty Quantification in Machine Learning Systems, Qualitative-Quantitative Uncertainty Integration to Machine Learning Systems, Compositional Data Analysis

Challenge area within FinTech we are trying to tackle

The primary challenge we are addressing is the issue of uncertainty in decision-making procedures within the FinTech industry. Many FinTech companies may face difficulties when they need to make critical decisions based on data that inherently includes a level of uncertainty. This could involve credit risk assessment, user feedback integration, or investment strategy formulation where the underlying data can be ambiguous or incomplete. Our focus is on enabling these companies to manage and interpret uncertain data effectively so they can make more informed decisions.

Solution and vision for the end product and concept

Our vision is to develop a software solution designed specifically for the FinTech sector. This software will be capable of collecting both data and the associated uncertainty related to that data. The unique aspect of our solution lies in its ability to process these dual layers of information—actual data points and their uncertainty margins—to deliver tailored, precise outcomes.

The end product will enable FinTech companies to integrate this enhanced decision-making tool into their existing systems. By doing so, they can improve the accuracy of their decisions, reduce risks associated with uncertainty, and optimise their strategies based on more reliable insights. Ultimately, our software aims to transform how FinTech companies handle uncertainty, turning potential vulnerabilities into strengths in their decision-making frameworks.

This approach not only helps in tactical day-to-day decision-making but also supports strategic planning and long-term business resilience in a sector that is continuously evolving and facing new challenges.

Property Portfolio Optimisation

Key expertise from UoN

Our team of researchers includes experts in economics, finance and the analysis of large data sets, and have had hands-on experience of new venture formation, management and successful exit. They have overseen and driven several collaborations between the University of Nottingham and industry partners. These collaborations have explored links between household financial decisions and financial well-being, in the domains of property, savings, investments, portfolio analysis, employment and income.

Challenge area within FinTech we are trying to tackle

We aim to improve financial decision-making by households, with the provision of tools that harness large data sets and apply relevant economic theory, AI, and/or financial risk models.

Such tools have the potential to bolster financial planning, alleviate financial stressors, and engender a more secure financial trajectory for UK households. There is likely to be concomitant use of such tools for the risk management function of banks and other providers of financial products.

Solution and vision

We have a number of active projects, foremost among these is the research into and development of tools that advise on property investments and borrowing.

For more information please contact us.

PRI

Key expertise from UoN

Our team of experts are leading this project in collaboration with a consortium of policymakers (Bank of England and Confederation of British Industry) and academic institutions (University of Bristol, University College London, University of Exeter, and Lancaster University). The project is an award of the UK Research and Innovation (UKRI) - Economic and Social Research Council (ESRC) with a grant size of around £300,000 for the period 2020-2022.

Challenge area within FinTech we are trying to tackle

It is challenging to predict SMEs’ solvency due to lack of data and shortage of sophisticated prediction models. Therefore, our research thoroughly employs advanced tools of Artificial Intelligence, Machine Learning, Deep learning and Big Data Analytics in analysing hundreds of variables from comprehensive public and proprietary data sources to develop a unique Pandemic Risk Index (PRI) that enhances the prediction of SMEs’ solvency during and post the Covid pandemic.

Solution and vision for the end product and concept

The particular takeaways to tease out from our research include: Our index (PRI) predicts SMEs’ solvency; Social, health and spatial aspects of Covid should be factored in the decision making of how much (emergency) funds to provide to SMEs; Foreign exposure of SMEs is a determinant factor of their covid-exposure due to severe disruptions in global supply chains.

Our vision is to develop a platform that utilises our index (PRI) to provide an interactive interface that enhances the prediction of SMEs’ solvency using firm level, sector level and regional data.

Predictive Machine Learning and Risk Insight

Key expertise from UoN

The Machine Learning in Science team are drawn from the School of Physics and Astronomy, specialising in Theoretical Physics and Machine Learning. They have specific expertise in applying complex mathematical and statistical techniques to real world problems, harnessing the capacity that modern Large Language Models offer to analyse textual and unstructured data.

Challenge area within FinTech we are trying to tackle

The team have brought together these techniques to identify anomalies in very large financial data sets, demonstrating the capability to analyse both quantitative and unstructured data sets when assessing risk, encompassing a range of use cases such as credit risk, customer retention and anomaly detection. As part of this development, the team are building novel solutions within the FCA Digital Sandbox programme to demonstrate our capability to identify hitherto occult risk factors in large and complex data sets pertinent to the Banking, Consumer Credit and Insurance sectors.

Solution and vision for the end product and concept

Once the technology has been showcased through the Sandbox, Infinity will seek to commercialise the respective applications.

Improving Readability through ML-enabled Linguistics

Key expertise from UoN

As part of a wider collaboration with colleagues from the Schools of English and Mathematical Sciences, the team have developed novel solutions to assess the 'readability' of complex documents, blending world-leading research in Linguistics with advanced statistical analysis and Machine Learning.

Challenge area within FinTech we are trying to tackle

The team are about to enter a proof-of-concept testing phase for the prototype application, which will be refined during the Autumn of 2024 with the ambition that a commercially viable product will be available by Spring 2025.

Solution and vision for the end product and concept

The application is applicable in most regulated sectors, particularly where compliance with the FCA Consumer Duty is required.

Inclusive Financial Technology Hub

Castle Meadow Campus
University of Nottingham
Nottingham
NG2 1AB

infinity@nottingham.ac.uk