UOB enhances anti-money laundering surveillance with some help from machine learning

UOB enhances anti-money laundering surveillance with some help from machine learning

Michelle Zhu
24/08/18, 04:32 pm

SINGAPORE (Aug 24): United Overseas Bank (UOB) says it has enhanced its anti-money laundering (AML) surveillance abilities with an integrated solution that enables the bank’s existing AML systems to make sharper, smarter and swifter detection of high-risk individuals & companies and suspicious activities.

This is essential given the volume, value and velocity of transactions that flow through UOB’s systems, says the bank in a press release on Friday.

The bank’s enhancement to its AML systems comes as a result of a recent collaboration with local regulatory technology (regtech) company Tookitaki Holding, says UOB, which saw the co-creation of a number of machine learning features for Tookitaki’s solution, Anti-Money Laundering Suite (AMLS).

Tookitaki is a graduate of the second accelerator programme of The FinLab, a joint venture between UOB and SGInnovate.  

After testing the effectiveness of AMLS over a six-month pilot, the bank has subsequently applied the solution for two of the four main processes within the AML framework, specifically ‘name screening’ and ‘transaction monitoring’.

In ‘name screening’, the bank identifies high-risk individuals and entities based on internal and external watch lists, enabling it to assess more accurately the risks involved to bank them, and to remain vigilant against AML activities by existing customers.

On the other hand, ‘transaction monitoring’ involves the identification and reporting of suspicious transactions of investigation.

UOB says it intends to progressively roll out AMLS to enhance the other two AML processes, namely ‘customer risk assessment’ and ‘sanctions screening’.

According to the bank, these new features enable AMLS – which is unique to the market as it can be applied to all processes within the AML framework – to conduct deeper and broader analyses of any set of data for greater accuracy.

When it spots a pattern of suspicious activity, the AMLS creates a smart rule and adds it to the AML typology library – thus enabling the machine learning models to detect similar patterns for future alerts.

This means the solution will continue to filter the number of false positives and hence contribute to more accurate tracking over time, all while boosting the productivity of UOB employees by enabling them to focus on other related tasks.

“The area of AML requires constant vigilance and continual enhancement to ensure that we stay on top of preventive, detective and enforcement measures. The use of RegTech such as Tookitaki’s AMLS enables us to augment our ability to identify actionable alerts and to minimise false positives. These sharpen the accuracy and effectiveness of our AML risk management,” says Victor Ngo, head of group compliance, UOB.

Singapore's housing dilemma: What are the broader implications?

SINGAPORE (Sept 17): In the past year and a half, the issue of the 99-year leases on Housing & Development Board (HDB) flats has triggered much debate. Now might be a good time to step back and examine what exactly has changed as a result of this debate and what the broader implications are. In fact, this issue could be a major game changer in many areas. Certainly, housing market dynamics will change, there could be shifts in savings behaviour and we are also likely to see major policy changes over time. These are big changes and it is important that Singaporeans have a clear understandin....

Najib to be charged tomorrow for RM2.6 bil in accounts

SINGAPORE (Sept 19): Malaysia’s former prime minister Najib Razak has been arrested today in connection to RM2.6 billion ($860 million) of 1MDB funds deposited into his personal account. According to a press release issued by the Malaysian Anti-Corruption Commission (MACC), Najib was arrested earlier today at 4.13pm at the MACC headquarters in Putrajaya. The former premier will be brought to the Kuala Lumpur Sessions Court tomorrow (Sept 20) at 3.00pm to face “several charges” under Section 23(1) of the MACC Act. The MACC also said that it will cooperate with the police to recor....

HNA explores sale of newly acquired CWT logistics unit

HONG KONG (Sept 19): China's HNA Group is in talks with banks to find a buyer for its CWT logistics unit, nine months after it acquired the Singaporean business in a US$1 billion  ($1.4 billion) deal, several people familiar with the matter told Reuters. See: CWT receives formal US$1 bil takeover offer from HNA See: HNA to face $3.1 bil liquidity crunch this quarter See: HNA's fire sale gets into full swing from Hong Kong to London See: HNA Group co-chairman Wang Jian dies after accident in France The sale, if completed, would be the latest in a series of divestments aimed ....