Crisis funds and loans put in place to help support businesses during the health emergency have become a prime target for cybercriminals. Neo4j’s Amy Hodler examines how graph technology could be a powerful weapon against these scams
Fraudsters will use any opportunity to siphon off funds illicitly, and the pandemic is proving no exception. With coronavirus moving rapidly across the world and locking down countries in its wake, cybercriminals have been quick to launch sophisticated methods to callously exploit the situation.
Cybercriminals have been fast to impersonate trusted organisations such as the World Health Organisation, which has itself seen a five-fold increase in cyberattacks since the start of the crisis.
The pandemic is opening the doors for fraudsters who are taking advantage of changes in normal business processes, controls and working conditions to carry out fraudulent activities. Security controls, for example, are often not as strong as normal due to the speed aid is required and the fact that many people are teleworking.
Cybercriminals are using fake or stolen identities to draw down governmental emergency funds. In France, for example, the Paris Prosecutor’s Office has launched an investigation into massive fraud of the country’s temporary unemployment scheme where fraudsters have drained €1.7 million. It is investigating potential international links to the fraud.
In a statement Paris Prosecutor Remy Heitz said that more than 1,740 fraudulent operations were discovered across the country on behalf of 1,069 different businesses asking for wire transfers to over 170 different bank accounts.
Can financial services’ practices help?
Aid departments and organisations should look to the mature practices of the financial services industry for a lead in combating fraud. Here firms repeatedly and meticulously check and compare transactional data to look for suspicious behaviour that may indicate an attack.
Like applications for financial aid for the impact of the coronavirus, malevolent actors look to defraud financial institutions using false identities when creating accounts and putting together loan applications. Personal data such as addresses, telephone numbers and emails are cleverly assembled to model assumed and phony identities.
A need for a different approach
One of the main reasons traditional approaches fall short is that most fraud detection systems are based on a relational database model where data is stored in predefined tables and columns. With large, unstructured data sets, relational databases swiftly reach their limits; queries turn out to be far too complex and response times lag. Banks and government authorities need the ability to follow a trail from one account to another, viewing a fraud network as a whole complete entity to work out how activities are linked.
Unlike relational databases, graph database technology not only represents individual items of data such as person, account number, home address, but also their relationships with one another such as how they are related. Any number of qualitative or quantitative properties can be assigned, showing complex relationships in an easy to understand way.
One of the best graph algorithms for fighting coronavirus cybercriminals is ‘PageRank’, which finds important nodes (objects) based on their relationships and interprets them using visualisation tools. For fraud detection in banking, the algorithm identifies important or influential customers who are featured in a large number of financial transactions. Nodes with a high PageRank Score can be illustrated using a visualisation tool so that they appear larger in the view and can be immediately picked up.
Another key algorithm is ‘Weakly Connected Components’, which works to reveal the hidden networks that form a fraud ring based on common identity features such as multiple applicants all residing at the same address. These hidden connections provide invaluable information when hunting down fraud.
Uncovering fraud rings with incredible accuracy
Cybercriminals are continually developing attack methods, sharing infrastructures to maximise their opportunities for success. Graph technology has the capacity to help stop advanced fraud scenarios in real time.
Graph databases can help future proof an organisation’s fraud prevention initiatives by enhancing insight based on data relationships and building connected intelligence.
The author is Director, Analytics and AI Program at Neo4j, the world’s leading graph database company, and co-author of Graph Algorithms: Practical Examples in Apache Spark & Neo4j, published by O’Reilly Media
