External Events

Emerging Tech take home a Prize at the FCA Hackathon

Share this post:

The FCA held a TechSprint (aka a hackathon) for three days in May for over 300 industry players to collaborate and explore ways to reduce financial crime. Four IBMers from  Emerging Technology, iX, and Software Group teamed up with a fin-tech, US regulator and bank to explore how to use natural language processing to analyse a broader set of unstructured data to reduce financial crime. We named ourselves Team Red Flag.
 
Ideas were flowing and the team often picked up and dropped ideas as we shared our knowledge with each other, complemented by very helpful injections by some of the floating SMEs at the event ranging from the legality, data sets and most importantly pain points today for professionals fighting financial crime.
 
We settle on one concept: using natural language processing to analyse information in SWIFT money transfer forms, focussing on the free text fields where payees provide information about a payment. 
 
Today’s rule-based transaction monitoring systems used by banks have limitations. They take up a lot of time as the systems are prone to false positive and false negative responses. This is very time intensive for people to sift to which is costly and it’s easy for criminals to circumvent. It’s estimated each year between $800 billion – $3 trillion is laundered around the world. 
 
The Concept
Red Flag takes in the big picture with natural language processing, to give meaning and context to payments for banks to block illicit payments and break the cycle of crime. Why bother? Well, today some state actors are bypassing sanctions and setting up mazes of front companies to conceal transactions and launder payments. The Velmur case saw North Korea illegally buy oil from Russia. If you drill down into the detailed transaction notes, mistakes were made by the criminals involved. At the time, no one flagged it was strange for a florist company to buy oil.
 
Our concept Red Flag does three things;
1.    Explore correlations between industries. So, how likely is a florist to transact with a oil company?
2.    We taught an AI to identify risk: it looked for well known known red flag phrases like ‘success fees’ or ‘advisory services’ in the payment instructions, but because we used AI rather than keyword searching we could identify risky phrases we had never seen before! just because they “sounded” risky.
3.    We Taught an second AI to read the free text of the transaction and make a guess at the industry it belonged to “Order for 200 roses” Probably a florist, “Install of new Routers & Wifi points” Probably a network & Comms company.
 
The prototype
We built an awesome visual prototype revealing what an AML analyst would see and then demonstrated a real transaction using data provided by the FCA that determined the industries of both the originator and beneficiary.We rated the transaction with a red flag score which was generated by a number of AI systems looking for suspicious phrases, unlikely industries transacting and then we created another AI system which learnt to predict the likelihood of the industries transacting together from the free text fields.  
 
Each of these  factors were combined to generate the overall Red Flag Score. Our aim was to support the AML analyst by helping them direct their focus with red flagged transactions.
 
Bigger Picture
In the future we’d include more data sets to be analysed including; corporate filings, news and media outlets, transaction instructions, regulation and advisories as well as location information. 
 
 
The Outcome
We were one of seventeen teams who pitched to around 400 Industry SMEs, Regulators and other hackers at EY’s offices in Canary Wharf. Team Red Flag were humbled and honoured to be awarded the Jump Prize, for the prototype with the most potential to transform and modernise the industry; to enable a great jump forward. 
 
Thanks to everyone who participated, EY for hosting and the FCA for running such an awesome event! 
More External Events stories

IBM Player Spotlight Built with IBM Watson

The Emerging Technology team within IBM Research UK, have been working with FOX Sports to enhance their viewer’s experience of watching ‘The Beautiful Game’. The IBM Player Spotlight Built with IBM Watson is an AI-powered tool built for the FOX Sports’ studio team to access complex statistical analyses using an easy-to-use conversational interface. With the […]

Continue reading

Playing with the Bloch Sphere

How we made a game of learning Qubits . In this article we will briefly show the tool we built while learning quantum computing; alas, this article is not about teaching these concepts. We will be covering these in a series of articles soon.     At the start of our journey we were learning […]

Continue reading

Raise the Game: Cognitive Analysis in Sports

A Clash of Worlds IBM Watson has been used to assist some of the world’s biggest companies and help tackle some of humanities greatest threats. Now, it’s being used to assist Leatherhead Football Club, a team that play in the 7th tier of English Football and are made of up delivery drivers, car salesmen and […]

Continue reading