With fraudulent activity on the rise, financial organisations need to keep up with their compliance requirements, avoiding fines and keeping themselves and their customers safe.
One method of fraud detection involves graph technology. Someone who knows all about this is Nik Vora, Vice President – APAC at Neo4j, a leader in graph technology. In this article, Nik describes how graph technology unearths data and connections to lead investigators to curb money laundering.
Identifying and stopping fraudulent activity is increasingly challenging with the rise of new technologies. In 2020, financial institutions in the Asia Pacific and globally racked up more than $10.6 billion in penalties. These penalties relate to non-compliance of Anti-Money Laundering (AML), Know Your Customer (KYC), MiFID (Markets in Financial Instruments Directive), and data privacy regulations. It’s clear that fraud detection is a high-stakes game. Businesses must recognise the necessity of being proactive and implementing preventative measures.
While traditional fraud detection methods play a crucial part in minimising these losses, criminals are getting more sophisticated, developing novel ways to elude discovery, acting either alone or as networks. A fraudster might create “new” synthetic identities – by stealing information from several people and mixing their social security numbers, addresses, phone numbers, and email addresses – later using those identities to open bank accounts, credit card accounts, or loans. To complicate things further, cyberspace gives them countless places to hide or disappear, making it incredibly difficult to find their trail, expose them, and eradicate these crimes.
Fraud Detection with Graphs
Sifting through reams of seemingly disparate data to find unusual or conflicting patterns that may lead to a potential money laundering transaction is an enormous task. Relational databases are not flexible enough to tackle changing, complex data structures; and, as such, struggle to compute relationships in the data.
Most fraud detection systems are based on traditional relational database models, which store data in separate, predefined tables and columns, requiring complex code to join relationships. While suitable for some business applications, these queries are time-consuming and a challenge to build and expensive to run on, resulting in slow response times. Such systems also put up false red flags, which can be detrimental to customer relations. To combat an issue such as money laundering with a solution like this one is costly and ineffective. While there is no way to get rid of the problem altogether, significant improvements are possible using graph database technology.
Graph databases such as Neo4j store data as “nodes” or relationships, which can be easily linked. Graphs interpret individual data such as ‘person,’ ‘account,’ ‘company,’ ‘address,’ and their relationships with one another, like ‘transacted with,’ or ‘registered at.’ Intuitive query languages like Cypher provide a simple semantic for detecting fraud rings in a graph, navigating connections in memory and in real-time to catch activities as they happen.
The graph data model below provides a visual representation of how data looks to the graph database, illustrating the 360-degree view and how one can recognise fraud rings by simply walking the graph.
Use Case: The International Consortium of Investigative Journalists
A great use case of what graphs can uncover is highlighted by the International Consortium of Investigative Journalists (ICIJ), the group behind the FinCEN Files, Panama Papers, and Paradise Papers. The ICIJ used graph technology, along with many other tools including Linkurious’ graph visualisation tool, to aid in exposing the world’s most extensive investigations of the rogue offshore finance industry. The group opted to go with graph technology because of its powerful and efficient capabilities in mapping complex financial connections and picking out irregularities. Graph technology played a significant part in joining the dots between discreet relationships, recouping more than $1.2 billion in resulting fines and back taxes. Listed in the Panama Papers leak were public officials and wealthy individuals, including Bollywood celebrities Amitabh Bachchan and Ajay Devgan, politicians Shishir Bajoria from West Bengal and former chief of the Delhi Lok Satta Party, Anurag Kejriwal, Delhi Land & Finance CEO Kushal Pal Singh, among hundreds of others.
The journalists behind these investigations did not have to be data scientists to work with a graph database; instead, they could simply click their way through the data while following connections.
The ICIJ has been nominated for a Nobel peace prize for the string of graph-powered investigations that started in 2011.
All graph database use cases share an underlying theme – the business’s need to understand connected data relationships better. Unlike most other ways of looking at data, graphs are designed to express relatedness. That’s why graphs have emerged as a faster, more accurate means of delivering invaluable insights, at scale and in real-time.
While traditional technologies continue to play a vital part in fraud prevention, they simply aren’t powerful enough on their own. As fraud attempts become increasingly complex, graph technology is a force multiplier for stopping these criminal activities in their tracks.
Discover how organisations are adding graph databases to uncover fraud rings and other sophisticated scams. Click to get your copy of Fraud Detection: Discovering Connections with Graph Database Technology.
The post Neo4j: Why Graph Technology Is the Key to Fraud Detection appeared first on The Fintech Times.