Financial fraud continues to be a pervasive issue despite technological advancements. Our project, the Financial Fraud Simulative Analytic Research Platform (FinFraudSIM), aims to address this by using 'Crime Script Analysis' to deconstruct fraud into sequential steps. This approach will identify points where AI can effectively intervene.

 

FinFraudSIM will simulate and visualize interactions between fraudsters and victims, allowing experts to develop strategies to disrupt fraud at various stages. This innovative platform integrates criminology, data science, behavioural analytics, and human-computer interaction, ensuring it is both cutting-edge and user-friendly. By engaging end users in the design process, we aim to create adaptable solutions to emerging fraud techniques, ultimately contributing to a more fraud-resilient society.

 

In this Proof-of-Concept Project we will carry out preliminary work for a larger-scale project to develop the FinFraudSIM platform. As such, we will identify gaps in existing financial fraud interventions, identify tasks where ML and AI can be integrated with Crime Script Analysis, explore the capabilities and limitations of ML and AI in this area, and finally, develop a non-functional mock-up of the FinFraudSIM platform.