Our pioneering scientists are recognised globally for their expertise and are tackling real-world problems with their innovative research.
Leading academics from our Computer Science Department are working with automation technology firm Caspian to further develop AI-driven automated Anti-Money Laundering Technology.
Investigating money laundering at global banks is a highly regulated domain, where it is essential to fully understand and explain the outcomes of the machine learning solutions.
Dr Al Moubayed is leading the collaboration between Durham University and Caspian through a Knowledge Transfer Partnership and Intensive Industrial Innovation Programme project to research the automatic narration of explainable decisions of machine learning models.
The project aims to automate the machine learning outputs in the language of the domain expert. This will allow the investigation teams in tier one banks to become more effective and efficient in investigating money laundering.
The project will build on Dr Al Moubayed’s expertise in natural language processing to develop a new approach to generate automatic summary analysis of the explainable output of machine learning models in a domain specific language.
Caspian will use this technology to automatically generate investigation reports directly in any language.
This will potentially result in a breakthrough in explainable machine learning and the further transparency of AI to the benefit of domain experts, regulatory bodies, and the public.
The research and resultant technology have the potential to be used in many other highly regulated sensitive domains including medicine, FinTech and pharmaceuticals.
Ian Blakemore, Knowledge Transfer Adviser, who has supported the project from inception, said: “This KTP is an example of knowledge transfer which improves the risk management of machine learning application in Fintech, improving explainable AI so that results are better understood, demystifying the technology for stakeholders. The potential for use in other sectors is particularly interesting.”
Knowledge Transfer Partnerships (KTPs) aim to help businesses to improve their competitiveness and productivity through the better use of knowledge, technology and skills within the UK knowledge base. This KTP project was funded by UKRI through Innovate UK.
The project is supporting a KTP associate, Dr Essel Ampomah, and a PhD student, Mr James Burton, who are advancing the research agenda and how it complements Caspian’s technology.