Granted Patent B2

US 12,566,825 · App. 17/570,410 · Granted Mar 3, 2026

Inventors: Daniel Vincent Safronoff (Washington, DC); Mahsa Chenari (Sterling, VA); Jessica Ya (Reston, VA)

Assignee: CAPITAL ONE SERVICES, LLC

G06F18/40G06F18/2178G06N3/08G06N20/20G06Q10/0639

Abstract

Claims (80)

1 . A system comprising:

one or more processors; and

a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:

receive first user interaction data associated with a user;

determine, using a first machine learning model (MLM), whether the first user interaction data exceeds a first predetermined threshold;

responsive to determining the first user interaction data exceeds the first predetermined threshold, assign a first training module to the user;

access a user profile associated with the user, the user profile comprising a plurality of training modules;

generate, using a second MLM, a first training plan based on the first training module and the plurality of training modules by:

assigning a first weight to each of one or more generation variables, and

determining a first order of the plurality of training modules based on the first weight of each of the one or more generation variables;

receive second user interaction data associated with the user;

determine a first efficacy level of the first training plan based on the second user interaction data and one or more validation variables;

dynamically update the first training plan using the second MLM based on the first efficacy level by:

assigning a second weight to each of the one or more validation variables, and

determining an updated first order of the plurality of training modules based on the second weight of each of the one or more validation variables; and

dynamically display in real-time the updated first training plan in the user profile with the updated first order comprising at least one of:

assigning one or more new training modules to the user,

eliminating one or more of the plurality of training modules, or

modifying content of one or more of the plurality of training modules.

2 . The system of claim 1 , wherein the first user interaction data and the second user interaction data comprise one or more of number of errors received by the user, degree of cursor movement, number of cursor clicks, time spent on a task, or combinations thereof.

3 . The system of claim 1 , wherein:

the first MLM comprises a neural network,

the first order is determined using a first cost function, and

the first efficacy level of the first training plan is determined using a second cost function.

4 . The system of claim 1 , wherein the instructions are further configured to cause the system to:

train the first MLM using one or more labels corresponding to one or more types of user interaction training data and indicating corresponding training thresholds, wherein

the first MLM determines whether the first user interaction data exceeds the first predetermined threshold by:

classifying at least a portion of the first user interaction data as a first type of one or more types,

determining the first predetermined threshold associated with the first type, and

determining whether the portion of the first user interaction data exceeds the first predetermined threshold.

5 . The system of claim 1 , wherein the one or more generation variables comprising one or more of whether a training module is required or optional, whether the user previously completed a training module, a length of time the user has been in a role, whether a training module may be overridden, user expertise, user experience, user demographics, or combinations thereof.

6 . The system of claim 1 , wherein determining the updated order of the plurality of training modules comprises modifying one or more of the first weight and the second weight.

7 . The system of claim 1 , wherein the one or more validation variables comprise one or more of a Customer Satisfaction (CSAT), an amount of time to complete the first training plan, user feedback, user surveys, user training flow, number of errors received by the user, or combinations thereof.

8 . The system of claim 1 , wherein dynamically updating the first training plan is further based on user completion of the first training plan.

9 . A system comprising:

one or more processors; and

a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:

receive first user interaction data associated with a user;

determine, using a first machine learning model (MLM), whether the first user interaction data exceeds a first predetermined threshold;

responsive to determining the first user interaction data exceeds the first predetermined threshold, assign a first training module to the user;

access a user profile associated with the user, the user profile comprising a plurality of training modules;

generate, using a second MLM, a first training plan based on the first training module and the plurality of training modules by:

assigning a first weight to each of one or more generation variables, and

determining a first order of the plurality of training modules based on the first weight of each of the one or more generation variables;

determine a first efficacy level of the first training plan based on one or more validation variables;

dynamically update the first training plan using the second MLM based on the first efficacy level by:

assigning a second weight to each of the one or more validation variables, and

determining an updated first order of the plurality of training modules based on the second weight of each of the one or more validation variables; and

update the user profile in real time to display the updated training plan with the updated first order by at least one of:

assigning one or more new training modules to the user,

eliminating one or more of the plurality of training modules, or

modifying content of one or more of the plurality of training modules.

10 . The system of claim 9 , wherein the first user interaction data comprises one or more of number of errors received by the user, degree of cursor movement, number of cursor clicks, time spent on a task, or combinations thereof.

11 . The system of claim 9 , wherein the one or more generation variables comprise one or more of whether a training module is required or optional, whether the user previously completed a training module, a length of time the user has been in a role, whether a training module may be overridden, user expertise, user experience, user demographics, or combinations thereof.

12 . The system of claim 9 , wherein the one or more validation variables comprise one or more of a Customer Satisfaction (CSAT), an amount of time to complete the first training plan, user feedback, user surveys, user training flow, number of errors received by the user, or combinations thereof.

13 . The system of claim 9 , wherein the instructions are further configured to cause the system to:

receive second user interaction data associated with the user.

14 . The system of claim 13 , wherein determining the first efficacy level of the first training plan is further based on the second user interaction data.

15 . A system comprising:

one or more processors; and

a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:

receive first user interaction data associated with a user;

determine, using a first machine learning model (MLM), whether the first user interaction data exceeds one or more predetermined thresholds;

responsive to determining the first user interaction data exceeds the one or more predetermined thresholds, assign one or more training modules to the user;

generate, using a second MLM, a first training plan based on the one or more training modules by:

assigning a first weight to each of one or more generation variables, and

determining a first order of the one or more of training modules based on the first weight of each of the one or more generation variables;

determine a first efficacy level of the first training plan based on one or more validation variables;

update the first training plan using the second MLM based on the first efficacy level by:

assigning a second weight to each of the one or more validation variables, and

determining an updated first order of the one or more of training modules based on the second weight of each of one or more validation variables; and

display in real-time the updated first training plan with the updated first order by at least one of:

assigning one or more new training modules to the user,

eliminating one or more of the training modules, or

modifying content of one or more of the training modules.

16 . The system of claim 15 , wherein the one or more generation variables comprise one or more of whether a training module is required or optional, whether the user previously completed a training module, a length of time the user has been in a role, whether a training module may be overridden, user expertise, user experience, user demographics, or combinations thereof.

17 . The system of claim 15 , wherein the one or more validation variables comprise one or more of a Customer Satisfaction (CSAT), an amount of time to complete the first training plan, user feedback, user surveys, user training flow, number of errors received by the user, or combinations thereof.

18 . The system of claim 15 , wherein the instructions are further configured to cause the system to:

receive second user interaction data associated with the user,

wherein determining the first efficacy level of the first training plan is further based on the second user interaction data.

Assignments (1)

Continuity (1)

References Cited (14)