INCREASING PERFORMANCE IN ANTI-MONEY LAUNDERING TRANSACTION MONITORING USING ARTIFICIAL INTELLIGENCE

• US 20190325528A1
• Filed: 04/24/2018
• Published: 10/24/2019
• Est. Priority Date: 04/24/2018
• Status: Abandoned Application

• Current Assignee: Brighterion Inc. (MasterCard Incorporated)
• Original Assignee: Brighterion Inc. (MasterCard Incorporated)
• Inventors: Adjaoute, Akli
• Application Number:
  US15/961,752
• Publication Number:
  US 20190325528A1
• Time in Patent Office: Days
• CPC Class Codes: G06F 18/2178 based on feedback of a supervisor
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/20 Ensemble learning
• CPC Class Codes: G06N 3/006 based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
• CPC Class Codes: G06N 3/048 Activation functions
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 5/041 Abduction
• CPC Class Codes: G06N 5/046 Forward inferencing; Production systems
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 20/4016 involving fraud or risk level assessment in transaction processing
• CPC Class Codes: G06Q 40/12 Accounting

Abstract

Provided is an artificial-intelligence based, electronic computer implemented system for generating an alert to a likelihood of money laundering activity within a financial environment, comprising at least one computer that includes both hardware and software components. The components form: a smart agent generating means for generating a smart agent for each entity capable of acting by itself or in concert with another in furtherance of money laundering activity; an updating means for updating each smart agent with transaction based data (financial and/or non-financial) associated therewith so that each smart agent models an individual entity behavior profile; a supervised learning model for a first-pass detection of potential money laundering activity; an unsupervised learning model for reducing false positive and enhancing detection of potential money laundering activity; and an alerting means for generating an alert to a likelihood of money laundering activity within the financial environment. Alternative systems are also provided.

Claims

1. An artificial-intelligence based, electronic computer implemented system for generating an alert to a likelihood of money laundering activity within a financial environment, comprising at least one computer that includes both hardware and software components, that together or individually form at least:

a smart agent generating means for generating a smart agent for each entity capable of acting by itself or in concert with another in furtherance of money laundering activity;

an updating means for updating each smart agent with transaction based data associated therewith so that each smart agent models an individual entity behavior profile;

a supervised learning model for a first-pass detection of potential money laundering activity;

an unsupervised learning model for reducing false positive and enhancing detection of potential money laundering activity; and

an alerting means for generating an alert to a likelihood of money laundering activity within the financial environment.

2. The system of claim 1, being a distributed system comprising a plurality of linked nodes.

3. The system of claim 1, wherein the supervised model comprises a plurality of models derived from patterns in a large set of past labeled behaviors

4. The system of claim 1, wherein the unsupervised model comprises a relevance sorting means.

5. The system of claim 4, wherein the relevance sorting means acts in conjunction with the updating means.

6. The system of claim 4, wherein the relevance s constructed to effect clustering, link analysis, associative learning techniques.

7. The system of claim 4, wherein the relevance sorting means allow for tracking of a characteristic selected from transaction volatility, entity interactions, and behavioral changes.

8. An artificial-intelligence based, electronic computer implemented system for complying with anti-money laundering rules and/or regulations within a financial environment, comprising at least one computer that includes both hardware and software components, that together or individually form at least:

a smart agent generating means for generating a smart agent for each entity capable of acting by itself or in concert with another in furtherance of money laundering activity;

an updating means for updating each smart agent with transaction based data associated therewith so that each smart agent models an individual entity behavior profile;

supervised and unsupervised learning models for detection of potential money laundering activity via input from one or more investigators; and

a behavior monitoring and/or analyzing means for detecting suspicious behavior as evidenced by application of the models to smart agent behavior within the financial environment.

9. The system of claim 8, being a distributed system comprising a plurality of linked nodes.

10. The system of claim 8, capable of multi-layer analysis to detect abnormal behavior.

11. The system of claim 10, wherein the multi-layer analysis includes analysis of a transaction layer, an individual or account, a business or organization layer, and a ring layer.

12. The system of claim 8, capable of carrying out multi-dimensional one-to-many and many-to-one behavioral analysis.

13. The system of claim 8, further comprising a watch list screening means for screening transactions against sanction and watch lists.

14. The system of claim 13, wherein the sanction and watch lists screening means phonetic and string matching algorithms leveraging fuzzy and flexible logic, as well as artificial-intelligence algorithms for unsupervised learning and clustering to reduce false positives.

15. An artificial-intelligence based, electronic computer implemented system for effecting case management in complying with anti-money laundering rules and/or regulations within a financial environment, comprising at least one computer that includes both hardware and software components, that together or individually form at least:

a smart agent generating means for generating a smart agent for each entity capable of acting by itself or in concert with another in furtherance of money laundering activity;

an updating means for updating each smart agent with transaction based data associated therewith so that each smart agent models an individual entity behavior profile;

supervised and unsupervised learning models for detection of potential money laundering activity; and

a contextual information analyzing means for detecting suspicious behavior under case management by application of the models to smart agent behavior within the financial environment.

16. The system of claim 15, being a distributed system comprising a plurality of linked nodes.

17. The system of claim 15, wherein the contextual information analyzing means is effective to offer one or more of the following:

(a) linking of customer relationships;

(b) comprehensive drill-in/drill-out capability;

(c) transaction alert investigation;

(d) work-flow and case management; and

(e) SARs/CTRs filing.