Machine Learning Rules vs. Models in Anti-Money Laundering Platforms
The value of money laundered globally each year is estimated to be 2-5% of global GDP. That falls between $800 billion and $2 trillion in USD. Machine learning models can help banks avoid the risks of ineffective AML solutions and help stop financial crime.
The Truth About Money Laundering
The estimated amount of money laundered globally in one year is 2 – 5% of global GDP, or $800 billion – $2 trillion in current US dollars. Machine learning can help banks avoid the repercussions of ineffective AML solutions and can actually help stop financial crime.
2–5% of global GDP | $800 – $2 Billion do not delete Trillion
Why Implement Machine Learning For AML?
- It creates more efficient and effective teams by automating case enrichment and prioritization for investigators.
- Automation significantly decreases the number of false positives generated.
- It allows for more accurate risk scoring.
In one instance, a bank reduced the time (taken to work alerts) from several weeks to a few seconds.
Rules-Only vs. Rules with Machine Learning Models
- Legacy AML systems provide high-volume, low-value alerts because they run on engines that only use rules. The overwhelming amount of false positives a rules-based system creates is akin to crying wolf.
- ML programs powered by machine learning often utilize both rules and models, not just rules. Using both rules and models dramatically reduces false positives, increases operational efficiency, and requires less maintenance.
How rules-based risk engines work
Rules-based risk engines work by using a set of mathematical conditions to determine what decisions to make.
Pros
- Analysts can quickly create and implement new rules in robust and innovative systems.
- A clear rule with specific calculations makes it easier to demonstrate to regulators why and when the system flagged the event as suspicious activity.
Cons
- Rules alone aren’t sufficient because they have too many limitations.
- Too complicated to understand context and dive deeper than formulas.
- Have fixed thresholds that criminals understand and purposely avoid.
- Only use YES/NO scenarios.
- Produce too many false positives.
- Require a great deal of manual effort to maintain.
- Have trouble detecting relationships between transactions.
How machine learning risk engines work
Machine learning for AML strengthens rules with models, which further reduces highvolume, low-value alerts
- Data science teams feed the machine massive amounts of historical data about known and suspected money laundering cases.
- Machine learning algorithms use the insights from these datasets to create statistical models, not deterministic rules.
- The machine learns what money laundering has looked like in the past and, equally important, what normal behavior the looks like as well.
- The machine predicts the risk of money laundering based on known and suspected money laundering cases or by referencing cases that were reported to the regulator.
Things to Note
- Machine learning models are only as good as their training data. The machine can’t learn without good, labeled data.
- Machine learning models take time to learn, making them slower to implement. But once they are deployed, machine learning makes up for that time by providing more accurate alerts.
- Machine learning saves your data science team countless hours they would have otherwise spent building and adjusting thousands of rules.