Moneying: RegTech and AML for the prevention of financial crimes

Artificial Intelligence solutions

Moneying RegTech and AML

Today, banks, insurance companies, financial intermediaries and a whole series of non-financial operators, such as gaming licensees, are required to comply with complex regulations governing digital financial flows.

The number of regulations in this area has increased tenfold in recent years, making it increasingly difficult for fintech and other organizations to maintain compliance.

The technological solutions that facilitate compliance with the regulatory requirements imposed on a subject constitute the RegTech, Regulatory Technology. In particular, RegTech’s software platforms use Artificial Intelligence techniques for Big Data analysis, to offer financial operators Analytics systems, which enable more precise monitoring of phenomena, and consequently allow for an increase in customer satisfaction and trust of the stakeholders.

In RegTech contexts, one of the most critical activities is related to anti-money laundering monitoring (AML), which determines a significant waste of time and resources for organizations.

Money-Laundering consists in giving a legal appearance to capital (goods, money or other utilities) of illicit origin, making it more difficult to identify it; Anti Money Laundering is a set of regulations and procedures created in order to prevent and control money laundering.

The explosion of digital payments that we have witnessed in recent years, accompanied by the emergence of increasingly pervasive mechanisms for the transfer of money, means that today it is essentially impossible to manage AML activities manually, and therefore that it is increasingly necessary make use of advanced technologies for the discovery of potentially harmful phenomena.

The Revelis Moneying platform

Revelis has developed Moneying, a platform that applies Machine Learning, Deep Learning and Text Analytics techniques for the analysis of financial transactions and customer communications.

Moneying provides for the automation of the following processes:

– evaluation of the customers’ risk profile (KYC);

– identification of AML phenomena and automatic reporting of the subjects involved to the UIF;

– calculation of risk indicators (Key Risk Indicator);

– claims management;

– management of reports from investigative bodies;

– sanctioning retail and business customers;

– access to customer information in complete and integrated form, through a file that is always available.

But how does Moneying work?

Moneying uses Machine Learning techniques to:

evaluate the risk profile of customers;

identify potential money laundering phenomena on the basis of information correlation algorithms;

analyze and identify suspicious transactions and phenomena relating to coalitions of subjects with anomalous behaviour;

– identify previously unknown laundering phenomena;

minimize false positives.

The characteristics of Moneying

Among the main features of Moneying it is possible to mention:

Extensibility: the set of AML indicators can be easily extended;

Risk Scoring: a risk profile is evaluated for each user which takes into account subjective and objective criteria;

Data Driven: the use of Big Data Analytics and Artificial Intelligence techniques allows the recalibration of the indicators in order to guarantee an exhaustive and more precise screening of the cases of anomaly;

Modularity: the platform offers distinct functional modules for AML, Risk Management, management of requests from Investigating Bodies and management of Complaints.

Moneying architecture

The functional architecture of Moneying is represented in the following figure: each functional area has its own implementation module which makes it easier to extend the platform and customize it.


The technical architecture consists of:

a back-end engine for the execution of Artificial Intelligence procedures, based on Rialto™, the Revelis technology for the implementation and execution of Big Data Analytics and Artificial Intelligence processes;

a web application that offers the services of analysis and verification of the data produced by the back-end.

Through the Rialto™ Desktop component, the data scientist carries out the analysis of the source data (transaction database), designs and implements the Artificial Intelligence procedures using the basic modules offered by Rialto™ and, in fact, customizing on the specific use case the functional and operational catalog offered by the platform.

Subsequently, the procedures are made executable on the Rialto™ Server component which will take care of automating their execution according to rules defined and configured for the specific case. 

The output of the procedures is stored in an ad hoc database (Big Data Warehouse) in order to be queried and shown to end users through the web app.


Example of a use case of the Moneying platform

Moneying has been used for a company of primary importance in the gaming sector focused on technological innovation as a front-line weapon in the fight against money laundering and in the fight to the financing of terrorism, to guarantee the effective identification of customers, the monitoring of suspicious transactions and reports to the UIF.

The verticalization on this scenario brings with it the following peculiarities:

all the requirements defined by the competent Authorities (Ministry of Internal Affairs, FIU, ADM) were preliminarily analyzed for the definition of the KRIs;

the KRIs have been scientifically defined, taking into account industry expertise;

the KRIs thresholds, defined according to statistical best practices, are dynamic as they are related to the trend of the data calculated by the System;

each player is assigned a Risk Score derived from the application of user profiling techniques that take into account subjective, objective and transactional elements, useful both in the on-boarding phase of new customers and for assigning priorities in the analysis of subjects suspected of money laundering;

each player is analyzed across all businesses and also considering information from external sources;

– system training allows the identification of Risk Models.

The benefits for the customer

Among the main advantages that have been mentioned by the customer himself after applying Moneying are:

scientific definition of KRIs;

reduction of false positives: it was possible through clustering and classification techniques identify a set of rules for the selection of anomalous movements, overcoming the approaches based on static thresholds currently used;


unitary and integrated view (internal and external sources) of the positions analyzed which allows for a weighted assessment of the actual risk;


automation of the generation of the .csv UIF track for SOS reporting with a saving of over 70% of the time currently used.



Characteristics, advantages that can be acquired and phenomena that can be monitored by this platform seen through this case of use of a company of primary importance in the gaming sector, demonstrates the possibility of applying the Moneying platform created by Revelis to the AML context regardless of the business of the company subject to the AML legislation.

Customer satisfaction is for Revelis the reward for the work done.

Author: Francesco Cupello

Do you want more information about Moneying? Contact us.