Developing next generation fraud prevention software

The University of Manchester's two-year partnership with law firm Kennedys has used modern data science methodologies to support fraud prevention.

The cost of fraud

Fraud is an increasing problem for insurance companies, with significant cost to both the industry and consumers. In 2023 alone, 84,400 fraudulent claims worth £1.1 billion were detected by the Association of British Insurers. Specialists in insurance fraud detection, Kennedys Law LLP, boast a number of products and services to support insurers when assessing claimants.

However, with fraudulent activities becoming more sophisticated and new techniques being developed, Kennedys needed to innovate to stay one step ahead. In partnership with the University’s Alliance Manchester Business School (AMBS), Kennedys looked to further develop their capabilities using modern data science techniques.

Dynamic solutions for shifting threats

Working alongside AMBS’s Decision and Cognitive Sciences Research Centre (DCSRC), Kennedys used their existing fraud expertise and international data sets to develop innovative machine learning solutions. They applied DCSRC’s evidential reasoning theory in the practical context of fraud detection. The resulting software uses data analytics techniques, semantic technologies, intelligent modelling methodologies and decision support systems to recognise strategies used by criminals that can be missed by humans. 

Creating long-lasting impact

Following the collaboration with AMBS, Kennedys saw its revenue boost by more than £15 million the following year in the UK. Richard West, Kennedys partner and lead of the firm’s Innovations Group, says: "The insight the work has given us, and the software we have developed, will greatly benefit our clients and our continuing work to help them manage claims in less time, at a lower cost and with better outcomes." Kennedys has also secured funding for another major KTP project to develop an emerging risk analytical tool for insurers.

AMBS’s partnership with Kennedys demonstrated the viability of applying the DCSRC’s evidential reasoning theories in a practical context, and built confidence in using these methodologies in other domains. Professor Jian-Bo Yang of the DCSRC says: "The same principles around transparent machine learning, artificial intelligence and evidence-based decision support can be applied to many other sectors such as law, healthcare, financial technology (FinTech) and engineering systems."