Granted Patent B2

US 12,737,738 · App. 18/752,931 · Granted Sep 15, 2026

Inventor: William Kievit (Stamford, CT)

Assignee: Synchrony Bank

G06Q20/10G06N20/00G06Q20/40

Abstract

Claims (71)

1 . A computer-implemented method, comprising:

accessing, by a computing device and from a database maintained by a payment instrument service, historical user spend data corresponding to one or more financial institution accounts and expenditures associated with a user, wherein the historical user spend data is organized according to a plurality of spend categories and is accessed in response to an application for a secured payment instrument;

processing, by the computing device, the historical user spend data and user credit data through an allowance generation machine learning system executing an allowance generation algorithm to identify an allowance grouping cluster, wherein the allowance grouping cluster is associated with a set of financial accounts corresponding to similarly situated users and to allowances allocated according to the plurality of spend categories, wherein the allowance generation algorithm is dynamically trained in real-time by processing an allowance dataset according to a set of allowance vectors corresponding to allowances allocated to a set of other secured payment instruments associated with other users and adjustments made to the allowances based on updated spend data associated with the other secured payment instruments;

identifying, by the computing device and based on the allowance grouping cluster, a line of credit and an initial allocation of a set of allowances for the secured payment instrument, wherein the initial allocation associated with the allowance grouping cluster is defined according to the plurality of spend categories;

issuing the secured payment instrument that is secured by the initial allocation and the one or more financial institution accounts;

dynamically updating, by the computing device, the allowance generation algorithm based on ongoing performance and updated spend data of different users associated with different secured payment instruments, and according to individual sets of allowances and graduation determinations associated with the different secured payment instruments;

processing, in real-time and by the computing device, new transaction data corresponding to the secured payment instrument through the updated allowance generation algorithm to generate a new allocation of the set of allowances for the secured payment instrument;

dynamically processing, by the computing device and as the new allocation is generated, the new transaction data through a graduation machine learning system executing a graduation algorithm to identify a graduation grouping cluster corresponding to a graduation determination for the secured payment instrument, wherein the graduation algorithm is trained using a graduation dataset according to a set of graduation vectors corresponding to the graduation determinations, and wherein the graduation dataset includes allowance data corresponding to different sets of allowances and adjustments made to the different sets of allowances obtained from the allowance generation machine learning system; and

providing, by the computing device and based on the graduation grouping cluster, an offer to graduate the secured payment instrument to an unsecured payment instrument.

2 . The computer-implemented method of claim 1 , further comprising:

obtaining new user credit data; and

adjusting the initial allocation according to the new user credit data.

3 . The computer-implemented method of claim 1 , further comprising:

updating a user interface associated with the user to present the initial allocation of the set of allowances;

detecting a set of modifications to the initial allocation of the set of allowances through the user interface; and

evaluating the set of modifications according to the historical user spend data, wherein when the set of modifications is accepted, the secured payment instrument is issued according to the set of modifications.

4 . The computer-implemented method of claim 1 , wherein the secured payment instrument is secured without a security deposit.

5 . The computer-implemented method of claim 1 , wherein the new allocation is generated in proportion to an increase in the line of credit associated with the secured payment instrument.

6 . The computer-implemented method of claim 1 , further comprising:

monitoring ongoing transactions associated with the secured payment instrument in real-time to obtain the new transaction data.

7 . The computer-implemented method of claim 1 , further comprising:

processing the graduation determinations associated with the different secured payment instruments and subsequent user performances according to the graduation determinations to update the graduation dataset; and

retraining the graduation algorithm using the updated graduation dataset.

8 . A system, comprising:

one or more processors; and

memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to:

access, from a database maintained by a payment instrument service, historical user spend data corresponding to one or more financial institution accounts and expenditures associated with a user, wherein the historical user spend data is organized according to a plurality of spend categories and is accessed in response to an application for a secured payment instrument;

process the historical user spend data and user credit data through an allowance generation machine learning system executing an allowance generation algorithm to identify an allowance grouping cluster, wherein the allowance grouping cluster is associated with a set of financial accounts corresponding to similarly situated users and to allowances allocated according to the plurality of spend categories, wherein the allowance generation algorithm is dynamically trained in real-time by processing an allowance dataset according to a set of allowance vectors corresponding to allowances allocated to a set of other secured payment instruments associated with other users and adjustments made to the allowances based on updated spend data associated with the other secured payment instruments;

identify, based on the allowance grouping cluster, a line of credit and an initial allocation of a set of allowances for the secured payment instrument, wherein the initial allocation associated with the allowance grouping cluster is defined according to the plurality of spend categories;

issue the secured payment instrument that is secured by the initial allocation and the one or more financial institution accounts;

dynamically update the allowance generation algorithm based on ongoing performance and updated spend data of different users associated with different secured payment instruments, and according to individual sets of allowances and graduation determinations associated with the different secured payment instruments;

process, in real-time, new transaction data corresponding to the secured payment instrument through the updated allowance generation algorithm to generate a new allocation of the set of allowances for the secured payment instrument;

dynamically process, as the new allocation is generated, the new transaction data through a graduation machine learning system executing a graduation algorithm to identify a graduation grouping cluster corresponding to a graduation determination for the secured payment instrument, wherein the graduation algorithm is trained using a graduation dataset according to a set of graduation vectors corresponding to the graduation determinations, and wherein the graduation dataset includes allowance data corresponding to different allowances obtained from the allowance generation machine learning system; and

provide, based on the graduation grouping cluster, an offer to graduate the secured payment instrument to an unsecured payment instrument.

9 . The system of claim 8 , wherein the instructions further cause the system to:

obtain new user credit data; and

adjust the initial allocation according to the new user credit data.

10 . The system of claim 8 , wherein the instructions further cause the system to:

update a user interface associated with the user to present the initial allocation of the set of allowances;

detect a set of modifications to the initial allocation of the set of allowances through the user interface; and

evaluate the set of modifications according to the historical user spend data, wherein when the set of modifications is accepted, the secured payment instrument is issued according to the set of modifications.

11 . The system of claim 8 , wherein the secured payment instrument is secured without a security deposit.

12 . The system of claim 8 , wherein the new allocation is generated in proportion to an increase in the line of credit associated with the secured payment instrument.

13 . The system of claim 8 , wherein the instructions further cause the system to:

monitor ongoing transactions associated with the secured payment instrument in real-time to obtain the new transaction data.

14 . The system of claim 8 , wherein the instructions further cause the system to:

process the graduation determinations associated with the different secured payment instruments and subsequent user performances according to the graduation determinations to update the graduation dataset; and

retrain the graduation algorithm using the updated graduation dataset.

15 . A non-transitory, computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to:

access, from a database maintained by a payment instrument service, historical user spend data corresponding to one or more financial institution accounts and expenditures associated with a user, wherein the historical user spend data is organized according to a plurality of spend categories and is accessed in response to an application for a secured payment instrument;

process the historical user spend data and user credit data through an allowance generation machine learning system executing an allowance generation algorithm to identify an allowance grouping cluster, wherein the allowance grouping cluster is associated with a set of financial accounts corresponding to similarly situated users and to allowances allocated according to the plurality of spend categories, wherein the allowance generation algorithm is dynamically trained in real-time by processing an allowance dataset according to a set of allowance vectors corresponding to allowances allocated to a set of other secured payment instruments associated with other users and adjustments made to the allowances based on updated spend data associated with the other secured payment instruments;

identify, based on the allowance grouping cluster, a line of credit and an initial allocation of a set of allowances for the secured payment instrument, wherein the initial allocation associated with the allowance grouping cluster is defined according to the plurality of spend categories;

issue the secured payment instrument that is secured by the initial allocation and the one or more financial institution accounts;

dynamically update the allowance generation algorithm based on ongoing performance and updated spend data of different users associated with different secured payment instruments, and according to individual sets of allowances and graduation determinations associated with the different secured payment instruments;

process, in real-time, new transaction data corresponding to the secured payment instrument through the updated allowance generation algorithm to generate a new allocation of the set of allowances for the secured payment instrument;

dynamically process, as the new allocation is generated, the new transaction data through a graduation machine learning system executing a graduation algorithm to identify a graduation grouping cluster corresponding to a graduation determination for the secured payment instrument, wherein the graduation algorithm is trained using a graduation dataset according to a set of graduation vectors corresponding to the graduation determinations, and wherein the graduation dataset includes allowance data corresponding to different sets of allowances and adjustments made to the different sets of allowances obtained from the allowance generation machine learning system; and

provide, based on the graduation grouping cluster, an offer to graduate the secured payment instrument to an unsecured payment instrument.

16 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:

obtain new user credit data; and

adjust the initial allocation according to the new user credit data.

17 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:

update a user interface associated with the user to present the initial allocation of the set of allowances;

detect a set of modifications to the initial allocation of the set of allowances through the user interface; and

evaluate the set of modifications according to the historical user spend data, wherein when the set of modifications is accepted, the secured payment instrument is issued according to the set of modifications.

18 . The non-transitory, computer-readable storage medium of claim 15 , wherein the secured payment instrument is secured without a security deposit.

19 . The non-transitory, computer-readable storage medium of claim 15 , wherein the new allocation is generated in proportion to an increase in the line of credit associated with the secured payment instrument.

20 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:

monitor ongoing transactions associated with the secured payment instrument in real-time to obtain the new transaction data.

21 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:

process the graduation determinations associated with the different secured payment instruments and subsequent user performances according to the graduation determinations to update the graduation dataset; and

retrain the graduation algorithm using the updated graduation dataset.

Assignments (1)

Continuity (2)

References Cited (22)