Updating a machine learning fraud model based on third party transaction information

Patent metadata

Field	Value
Publication number	US-10672005-B1
Application number	US-201916279723-A
Country	US
Kind code	B1
Filing date	Feb 19, 2019
Priority date	Feb 19, 2019
Publication date	Jun 2, 2020
Grant date	Jun 2, 2020

Abstract

A device receives first transaction information associated with a first transaction, and a first transaction account utilized for the first transaction and associated with a first financial institution. The device determines, based on a fraud model, that the first transaction is to be denied due to potential fraud associated with the first transaction account and receives second transaction information associated with a second transaction, and a second transaction account utilized for the second transaction and associated with a second financial institution. The device processes the first transaction information and the second transaction information, with a matching model, to determine whether the first transaction information matches the second transaction information and determines that the first transaction was incorrectly denied when the first transaction information matches the second transaction information within a predetermined threshold. The device performs one or more actions based on determining that the first transaction was incorrectly denied.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a device, first transaction information associated with a first transaction based on a first transaction card being utilized to conduct the first transaction,
wherein the first transaction information includes information associated with:
the first transaction, and
a first transaction account utilized for the first transaction and associated with a first financial institution;

determining, by the device and based on a fraud model, that the first transaction is to be denied due to a potential fraud associated with the first transaction card;
receiving, by the device, second transaction information associated with a second transaction based on a user device or a second transaction card being utilized to conduct the second transaction,
wherein the second transaction information includes information associated with:
the second transaction, and
a second transaction account utilized for the second transaction and associated with a second financial institution;

processing, by the device, the first transaction information and the second transaction information, with a matching model, to determine whether the first transaction information matches the second transaction information;
determining, by the device, that the first transaction was incorrectly denied, due to the potential fraud associated with the first transaction account, when the first transaction information matches the second transaction information within a predetermined threshold; and
performing, by the device, one or more actions based on determining that the first transaction was incorrectly denied,
wherein performing the one or more actions includes:
removing, based on determining that the first transaction was incorrectly denied after the user device or the second transaction card is utilized to conduct the second transaction, a fraud lock on the first transaction card to allow the first transaction card to be utilized for future transactions.

2. The method of claim 1 , further comprising:
processing the second transaction information to generate processed second transaction information with a same format as the first transaction information.

3. The method of claim 1 , wherein performing the one or more actions comprises:
updating training data for the fraud model, based on parameters of the fraud model that caused the first transaction to be incorrectly denied, to generate updated training data; and
retraining the fraud model with the updated training data.

4. The method of claim 1 , wherein performing the one or more actions further comprises one or more of:
providing a customer service communication to the user device;
providing reward points to an account associated with a user of the user device; or
providing, to the user device, a communication requesting confirmation that the first transaction was incorrectly denied.

5. The method of claim 1 , wherein performing the one or more actions further comprises one or more of:
providing a promotion to the user device;
increasing a spending limit associated with the first transaction account;
reducing an interest rate associated with the first transaction account; or
increasing a cash back offer associated with the first transaction account.

6. The method of claim 1 , wherein processing the first transaction information and the second transaction information with the matching model comprises one or more of:
comparing a first amount associated with the first transaction and a second amount associated with the second transaction;
comparing first merchant information associated with the first transaction and second merchant information associated with the second transaction;
comparing first location information associated with the first transaction and second location information associated with the second transaction;
comparing first source information indicating whether the first transaction occurred online or at a first physical location and second source information indicating whether the second transaction occurred online or at a second physical location;
comparing first time information associated with the first transaction and second time information associated with the second transaction; or
comparing first date information associated with the first transaction and second date information associated with the second transaction.

7. The method of claim 1 , wherein receiving the second transaction information associated with the second transaction comprises:
receiving the second transaction information from the second financial institution, from a third party, or via scraping.

8. A device, comprising:
one or more memories; and
one or more processors, communicatively coupled to the one or more memories, configured to:
receive first transaction information associated with a first transaction based on a first transaction card being utilized to conduct the first transaction,
wherein the first transaction information includes information associated with:
the first transaction, and
a first transaction account utilized for the first transaction and associated with a first financial institution;

deny, based on a fraud model, the first transaction due to a potential fraud associated with the first transaction card;
receive second transaction information associated with a second transaction based on a user device or a second transaction card being utilized to conduct the second transaction,
wherein the second transaction information includes information associated with:
the second transaction, and
a second transaction account utilized for the second transaction and associated with a second financial institution;

process the second transaction information to generate processed second transaction information with substantially a same format as the first transaction information;
process the first transaction information and the processed second transaction information, with a matching model, to determine whether the first transaction information matches the processed second transaction information;
determine that the first transaction was incorrectly denied, due to the potential fraud associated with the first transaction account, when the first transaction information matches the processed second transaction information within a predetermined threshold; and
perform one or more actions based on determining that the first transaction was incorrectly denied,
wherein, when performing the one or more actions, the one or more processors are configured to:
remove, based on determining that the first transaction was incorrectly denied after the user device or the second transaction card is utilized to conduct the second transaction, a fraud lock on the first transaction card to allow the first transaction card to be utilized for future transactions.

9. The device of claim 8 , wherein the matching model includes a fuzzy matching-based machine learning model.

10. The device of claim 8 , wherein, when performing the one or more actions, the one or more processors are configured to:
update training data for the fraud model, based on parameters of the fraud model that caused the first transaction to be incorrectly denied, to generate updated training data; and
retrain the fraud model with the updated training data.

11. The device of claim 8 , wherein, when performing the one or more actions, the one or more processors are further configured to one or more of:
provide a customer service communication to the user device a;
provide reward points to an account associated with a user of the user device;
provide, to the user device, a communication requesting confirma

Assignees

• Capital One Services Llc

Inventors

• Hart Colin
• Edwards Joshua
• Perezleon Francisco
• Johnson Molly
• Newman Kaitlin
• Wu Angelina
• Ji Jason

Classifications

• G06Q20/4016 Primary

  involving fraud or risk level assessment in transaction processing · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• H04L63/20

  for managing network security; network security policies in general (filtering policies H04L63/0227) · CPC title

• H04L63/102

  Entity profiles · CPC title

• G06Q20/401

  Transaction verification · CPC title