PhD Study : Network Machine Learning Approach to Financial Crime Detection

Summary

In recent years, financial crime has attracted a great deal of concern and attention. Financial crime including fraud and money laundering is happening at an unprecedented scale. The United Nations Office on Drugs and Crime (UNODC) estimates that between 2 and 5% of global GDP is laundered each year. That’s between EUR 715 billion and 1.87 trillion each year with less than 1% of this being caught. To address these challenges, the companies within the financial sector are making substantial investments in technology to help them make use of their internal and external data. It is essential that institutions know their customer and can validate their identify.

Critical to this are effective data wrangling, entity resolution, and linkage methods. Producing a golden record / single customer view requires the integration of large volumes of counter-party data from multiple sources (including internal and external sources). These data are often messy containing many duplications and issues in terms of spelling and address differences. Uncovering the networks and modelling connections within these data can result in models used to detect patterns indicative of financial crime, sanction list connections and other risks. These are important for retail, investment and commercial banking as well as capital markets use cases. Statistical approach and machine learning methods have been applied to detect the crime. Most of current approaches are based of supervising learning. However, the accuracy of the detection remains challenging.

This project aims to develop network based machine learning algorithms to detect financial crime. The project will focus on the extraction of network links between for example entities using the platform and integration with other data. Network/link analysis will be performed to identify patterns among identities along with the graphical representation the resulting networks. The network analysis algorithms developed will be integrated and used to extend the Datactics platform which provides data matching, cleansing and reformatting capabilities to clients in the FinTech and GovTech sectors.

Drawing on knowledge from other domains, the PhD Researcher will develop radical innovations which show promise for integration into existing finance AI detection system. The PhD Researcher will have the opportunity to interact with researchers from the collaborating company Datatics. These interactions will provide the student with a strong set of research and technical skills to complete a successful and expeditious research program.

Reference

[1] IBM, Fighting financial crime with AI, available at https://www.ibm.com/downloads/cas/WKLQKD3W [1] SADGALI, I., SAEL, N., & BENABBOU, F. (2019). Performance of machine learning techniques in the detection of financial frauds. Procedia computer science, 148, 45-54.

[2] Weber, M., Domeniconi, G., Chen, J., Weidele, D. K. I., Bellei, C., Robinson, T., & Leiserson, C. E. (2019). Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics. arXiv preprint arXiv:1908.02591.

Essential criteria

Applicants should hold, or expect to obtain, a First or Upper Second Class Honours Degree in a subject relevant to the proposed area of study.

We may also consider applications from those who hold equivalent qualifications, for example, a Lower Second Class Honours Degree plus a Master’s Degree with Distinction.

In exceptional circumstances, the University may consider a portfolio of evidence from applicants who have appropriate professional experience which is equivalent to the learning outcomes of an Honours degree in lieu of academic qualifications.

• Sound understanding of subject area as evidenced by a comprehensive research proposal

Desirable Criteria

If the University receives a large number of applicants for the project, the following desirable criteria may be applied to shortlist applicants for interview.

• Publications record appropriate to career stage
• A comprehensive and articulate personal statement
• Applicants will be shortlisted if they have an average of 75% or greater in a first (honours) degree (or a GPA of 8.75/10). For applicants with a first degree average in the range of 70% to 74% (GPA 3.3): If they are undertaking an Masters, then the average of their first degree marks and their Masters marks will be used for shortlisting.

Funding and eligibility

This project is funded by:

• DfE CAST Award in collaboration with Datatics Ltd

The scholarship will cover tuition fees at the Home rate and a maintenance allowance of £16,009 per annum for three years. EU applicants will only be eligible for the fees component of the studentship (no maintenance award is provided). For non-EU nationals the candidate must be 'settled' in the UK. This scholarship also comes with £900 per annum for three years as a research training support grant (RTSG) allocation to help support the PhD Researcher.