Introduction
The Anti Money Laundering solution has modules for transaction monitoring and name screening. The typical AML system has a large number of false alerts. They are measured for their efficiency in detecting the laundering cases. Identification of the cases is tougher if the technology is used by the violators. Banks can gain by cutting down the regulatory risks and damages due to compliance issues. The platform helps in cutting down the false alerts and minimizing risk through true positives.
The platform is based on machine learning and artificial intelligence solution which matches the case based on the specific rules and compliance policies. The platform improves the quality and efficiency of the enterprise and cuts down the cost of reconciliation. Match cases percentage and exception count is the measure of success of the solution. The rules are modeled based on compliance policies and rules. Machine Learning algorithms detect patterns based on the violation scenarios and exceptions & breaks are identified.
The platform can be deployed on both the cloud and on-premise. The deployment configuration typically has CPU clusters, storage, load balancers, and other deployment nodes. It can be also deployed on distributed parallel nodes. Spark and Hadoop Big data stacks can be deployed with an in-memory database, Enterprise Service Bus, or any other middleware to have analytics and machine learning models running based on the big data. The solution can be horizontally scalable to move hand-in-hand with big data sets. With In-Memory analytics, the platform can run machine learning for critical business use cases. The architecture will have the security mechanisms to facilitate single sign-on, authorization, auditing, logging, and monitoring. The deployed services can work on secured data in-transit and at-rest. Security architecture can have a role-based authorization, granular permissions model, and administration management. System Architecture has to enforce Access Control lists starting from the application layer to all the way down to the Hadoop layer. The cloud-based deployment can have the load balancer, web server, and database on EC2 nodes, CDN, and content storage on S3. The static files can be accessed through the content delivery network. The dynamic content can be accessed through the webserver.
The platform handles the following use cases related to anti-money laundering:
- Transferring money offshore to tax havens
- Allows criminals to transfer and share cash anonymously.
- Inflated prices used to funnel in cash to legal business
- Money is offshored and legally returned as an investment.
- Criminals transfer money from all accounts of the bank
- Unregistered employees with no paperwork getting cash salaries
- Legalize unreported assets
From the compliance side, the platform will have the following features:
- Monitoring and Reporting
- Customer Due diligence
- Know your customer
Blockchain technology is evolving to handle AML platform features. The future AML platform will be on a blockchain.