Billions of dollars are lost every year due to money laundering in Australia, with the global figure potentially reaching trillions.
Money laundering is often carried out by creating shell companies that have no physical presence or active business operations.
These companies are frequently linked to a wide range of illicit activities such as bribery, corruption, tax evasion, terrorist financing and cybercrime.
Now global money laundering can be tracked and identified much faster, giving tax officials and police an advantage.
Professor Kuldeep Kumar from the Centre for Data Analytics at Bond University has created a new data analytics model, detailed in a paper published in the Journal of Economic Criminology.
The model correctly identifies illicit shell companies with an accuracy of 88-97 percent.
Prof Kumar worked on the project alongside fellow researchers Miland Tiwari from the Australian Graduate School of Policing and Security and Dr Adrian Gepp from the Bond Business School.
The International Monetary Fund estimates money laundering accounts for 2-5 percent of the world’s annual gross domestic product or about $US1.5 trillion.
In Australia the figure is $10–$15 billion per year.
“These illegal activities (money laundering) constitute a profound threat to global economies as they enable criminal enterprises to undermine the integrity of the financial system and foster a shadow economy,” Prof Kumar said.
The new model will particularly benefit government officials and compliance professionals, including accountants, tax officials and anti-corruption agencies.
It uses a hybrid approach combining graph analytics and supervised machine learning, and was trained on publicly available data on shell companies identified in 804 corruption cases.
“From a network perspective, money laundering exploits a web of disguised relationships, and unravelling this complex network of connections is suited to graph analytics,” Prof Kumar said.
“This model can facilitate analysis of hidden networks enabling investigators to detect patterns, infer relationships and identify suspicious activities.”
Prof Kumar said previous work to detect illicit activities such as money laundering had focused on banking transactions which may not be publicly available and increase the burden of tracking an audit trail.
“Our research highlights the growing interest in the use of data-driven software tools to increase the detection of money laundering activities.”