Utilizing machine learning to generate recommendations for a transaction based on loyalty credits and stored-value cards

Patent metadata

Field	Value
Publication number	US-2019180255-A1
Application number	US-201715839374-A
Country	US
Kind code	A1
Filing date	Dec 12, 2017
Priority date	Dec 12, 2017
Publication date	Jun 13, 2019
Grant date	—

Abstract

A device receives a first set of information that relates to bank accounts associated with users, and receives a second set of information that relates to loyalty credits associated with the users. The device receives a third set of information that relates to stored-value cards associated with the users, and trains a model based on the first, second, and third sets of information. The device receives, from a client device, a request for a transaction, and utilizes the trained model to generate recommendations. The device provides, to the client device, the recommendations and a request for transaction information, and receives, from the client device, the transaction information, where the transaction information includes account information, loyalty credits information, and stored-value card information. The device determines transaction terms based on the account information, the loyalty credits information, and the stored-value card information, and provides the transaction terms to the client device.

First claim

1 . A device, comprising:
one or more memories; a communication interface to communicate with a first group of servers, a second group of servers, and a third group of servers; a machine learning component; and one or more processors, communicatively coupled to the one or more memories, to:
receive, via the communication interface, a first set of information from the first group of servers,
the first set of information relating to financial accounts associated with a plurality of users,
the first set of information relating to prior transaction information associated with the plurality of users;

receive, via the communication interface, a second set of information from the second group of servers,
the second set of information relating to loyalty credits associated with the plurality of users,
the second set of information relating to prior transaction information associated with the plurality of users;

receive, via the communication interface, a third set of information from the third group of servers,
the third set of information relating to stored-value cards associated with the plurality of users,
the third set of information relating to prior transaction information associated with the plurality of users;

store the prior transaction information associated with the first set of information, the second set of information, and the third set of information in a data structure for further processing;
apply one or more security techniques to protect the prior transaction information while the prior transaction information is being stored;
train a model, via the machine learning component, based on the first set of information, the second set of information, and the third set of information,
the model being trained to determine patterns associated with respective prior transaction information related to the first set of information, the second set of information, and the third set of information,
the model being a collaborative filtering model filtering the patterns associated with the prior transaction information associated with the plurality of users;

receive, from a client device associated with a user, a request for a transaction;
utilize the trained model to generate one or more recommendations associated with the transaction;
provide, to the client device, the one or more recommendations and a request for transaction information associated with the user;
receive, from the client device, the transaction information based on the request for the transaction information,
the transaction information including:
account information associated with a financial account of the user, and
at least one of;
loyalty credits information identifying loyalty credits associated with the user, or
stored-value card information identifying a stored-value card associated with the user;

determine transaction terms for the transaction based on validation of the transaction information associated with the user;
provide information identifying the transaction terms to the client device;
receive information indicating at least one of a quantity of the loyalty credits or an amount of the stored-value card to apply to a particular term of the transaction terms;
modify the particular term based on the information indicating the at least one of the quantity of the loyalty credits or the amount of the stored-value card to apply to the particular term; and
provide the modified particular term to the client device.

2 . The device of claim 1 , where the one or more processors are further to:
receive an acceptance or a rejection of the transaction terms from the client device.

3 . (canceled)

4 . The device of claim 1 , where the one or more processors are further to:
receive an acceptance or a rejection of the particular term from the client device.

5 . The device of claim 1 , where the particular term includes one of:
a down payment for the transaction, an insurance fee for the transaction, a tax fee for the transaction, or a closing fee for the transaction.

6 . The device of claim 1 , where the one or more recommendations includes one or more of:
a recommendation for a particular product, a recommendation for a particular service, a recommendation for a particular stored-value card, or a recommendation for particular loyalty credits.

7 . The device of claim 1 , where the transaction includes at least one of:
a vehicle loan, a mortgage loan, a payday loan, an appliance loan, a home equity loan, a student loan, a personal loan, or a small business loan.

8 . A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors to:
receive a first set of information from a first group of servers,
the first set of information relating to financial accounts associated with a plurality of users,
the first set of information relating to prior transaction information associated with the plurality of users;

receive a second set of information from a second group of servers,
the second set of information relating to loyalty credits associated with the plurality of users,
the second set of information relating to prior transaction information associated with the plurality of users;

receive a third set of information from a third group of servers,
the third set of information relating to stored-value cards associated with the plurality of users,
the third set of information relating to prior transaction information associated with the plurality of users;

store the prior transaction information associated with the first set of information, the second set of information, and the third set of information in a data structure for further processing;
apply one or more security techniques to protect the prior transaction information while the prior transaction information is being stored;
train, via a machine learning component, a model based on the first set of information, the second set of information, and the third set of information,
the model being trained to determine patterns associated with respective prior transaction information related to the first set of information, the second set of information, and the third set of information,
the model being a collaborative filtering model filtering the patterns associated with the prior transaction information associated with the plurality of users;

receive, from a client device associated with a user, a request for a transaction;
utilize the trained model to generate one or more recommendations associated with the transaction;
provide, to the client device, the one or more recommendations and a request for transaction information associated with the user;
receive, from the client device, the transaction information based on the one or more recommendations and based on the request for the transaction information,
the transaction information including:
account information associated with a financial account of the user, and
loyalty credits information identifying loyalty credits associated with the user;

determine transaction terms for the transaction based on validation of the transaction information associated with the user;
provide information identifying the transaction terms to the client device;
receive information indicating at least one of a quantity of the loyalty credits or an amount of the stored-value card to apply to a particular term of the transaction terms;
modify the particular term based on the information indicating the at least one of the quantity of the loyalty credits or the amount of the stored-value card to apply to the particular term; and
provide the modified particular t

Assignees

• Capital One Services Llc

Inventors

• Deshpande Amit
• Chikkaveerappa Satish
• Venuraju Mithra Kosur

Classifications

• G06Q40/03

  Credit; Loans; Processing thereof · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06Q20/387

  Payment using discounts or coupons (relating to marketing G06Q30/02) · CPC title

• G06Q30/04

  Billing or invoicing · CPC title

• G06N20/00

  Machine learning · CPC title