Granted Patent B2

US 12,356,042 · App. 18/514,920 · Granted Jul 8, 2025

Inventors: Donna Maria Welch (Bluffton, SC); Sudhakar Balu (Chennai, IN); Srinivasa Dhanwada (Hyderabad, IN); Siva Kumar Paini (Hyderabad, IN)

Assignee: Bank of America Corporation

H04N21/47202H04L51/02H04L67/63H04N21/4431H04N21/47217G06F16/951G06F40/30H04L49/90

Abstract

Claims (104)

1. A system comprising:

a memory configured to store:

a set of videos, each video of the set of videos associated with an application of a set of applications and depicting a process performed using the application;

a hardware processor communicatively coupled to the memory, the hardware processor configured to:

for each video of the set of videos:

generate metadata associated with the process depicted in the video; and

store the metadata within a collection of metadata stored in the memory; receive a request for information from a user;

in response to receiving the request for the information:

identify a first application of the set of applications for use in generating a response to the request;

generate, based on the collection of metadata stored in the memory, instructions for interfacing with the first application to generate the response to the request;

execute the instructions to interface with the first application to generate the response to the request; and

transmit the response to the user;

wherein, in response to receiving the request for information, the hardware processor is further configured to:

store the request for information at an end of a queue; and

monitor a position of the request within the queue;

wherein generating the response to the request is performed in response to determining that the request is at a top of the queue;

wherein, in response to storing the request for information at the end of the queue, the hardware processor is further configured to:

collect values for a set of parameters associated with the system;

determine, based on the request for the information and the collected values for the set of parameters, an estimate of a time to generate the response to the request; and

transmit, to the user, the estimate of the time to generate the response to the request;

wherein:

the memory is further configured to store:

training data comprising a plurality of historical requests, wherein for each historical request of the plurality of historical requests, the training data further comprises a set of historical parameters associated with the historical request and a historical time taken to respond to the historical request; and

a machine learning algorithm; and

the hardware processor is further configured to train the machine learning algorithm, based on the training data, to determine the estimate of the time to generate the response to the request; and

determining the estimate of the time to generate the response to the request comprises executing the machine learning algorithm.

2. The system of claim 1 , wherein the instructions for interfacing with the first application comprise at least one of:

a custom API for interfacing with the application; and

a set of headless browser steps for interfacing with the application.

3. The system of claim 1 , wherein:

the memory is further configured to store a plurality of sets of instructions, wherein each set of instructions of the plurality of sets of instructions:

is associated with a particular process depicted in a particular video of the set of videos;

comprises instructions for interfacing with the application associated with the particular video; and

was generated based on the metadata associated with the particular process depicted in the particular video; and

generating the instructions for interfacing with the first application comprises identifying the instructions for interfacing with the first application within the plurality of sets of instructions stored in the memory.

4. The system of claim 1 , wherein:

the memory is further configured to store a plurality of sets of instructions, wherein each set of instructions of the plurality of sets of instructions:

is associated with a particular process depicted in a particular video of the set of videos;

comprises instructions for interfacing with the application associated with the particular video; and

was generated based on the metadata associated with the particular process depicted in the particular video; and

generating the instructions for interfacing with the first application comprises executing a machine learning algorithm trained to generate the instructions for interfacing with the first application, based on the plurality of sets of instructions stored in the memory.

5. A method comprising:

recording a set of videos, each video of the set of videos associated with an application of a set of applications and depicting a process performed using the application;

for each video of the set of videos:

generating metadata associated with the process depicted in the video; and

storing the metadata within a collection of metadata;

receiving a request for information from a user;

in response to receiving the request for the information:

identifying a first application of the set of applications for use in generating a response to the request;

generating, based on the collection of metadata, instructions for interfacing with the first application to generate the response to the request;

executing the instructions to interface with the first application to generate the response to the request; and

transmitting the response to the user;

wherein, in response to receiving the request for information, the method further comprises:

storing the request for information at an end of a queue; and

monitoring a position of the request within the queue;

wherein, generating the response to the request is performed in response to determining that the request is at a top of the queue;

wherein, in response to storing the request for information at the end of the queue, the method further comprises:

collecting values for a set of parameters associated with the system;

determining, based on the request for the information and the collected values for the set of parameters, an estimate of a time to generate the response to the request; and

transmitting, to the user, the estimate of the time to generate the response to the request;

wherein:

the method further comprises training a machine learning algorithm, based on training data, to determine the estimate of the time to generate the response to the request, the training data comprising a plurality of historical requests, wherein for each historical request of the plurality of historical requests, the training data further comprises a set of historical parameters associated with the historical request and a historical time taken to respond to the historical request; and

determining the estimate of the time to generate the response to the request comprises executing the machine learning algorithm.

6. The method of claim 5 , wherein the instructions for interfacing with the first application comprise at least one of:

a custom API for interfacing with the application; and

a set of headless browser steps for interfacing with the application.

7. The method of claim 5 , wherein generating the instructions for interfacing with the first application comprises identifying the instructions for interfacing with the first application within a plurality of sets of instructions stored in a memory, wherein each set of instructions of the plurality of sets of instructions:

is associated with a particular process depicted in a particular video of the set of videos;

comprises instructions for interfacing with the application associated with the particular video; and

was generated based on the metadata associated with the particular process depicted in the particular video.

8. The method of claim 5 , wherein generating the instructions for interfacing with the first application comprises executing a machine learning algorithm trained to generate the instructions for interfacing with the first application, based on a plurality of sets of instructions stored in a memory, wherein each set of instructions of the plurality of sets of instructions:

is associated with a particular process depicted in a particular video of the set of videos;

comprises instructions for interfacing with the application associated with the particular video; and

was generated based on the metadata associated with the particular process depicted in the particular video.

9. A non-transitory computer-readable medium comprising instructions that, when executed by a hardware processor, cause the processor to:

record a set of videos, each video of the set of videos associated with an application of a set of applications and depicting a process performed using the application;

for each video of the set of videos:

generate metadata associated with the process depicted in the video; and

store the metadata within a collection of metadata;

receive a request for information from a user;

in response to receiving the request for the information:

identify a first application of the set of applications for use in generating a response to the request;

generate, based on the collection of metadata, instructions for interfacing with the first application to generate the response to the request;

execute the instructions to interface with the first application to generate the response to the request; and

transmit the response to the user;

wherein, in response to receiving the request for information, the instructions, when executed by the hardware processor, further cause the processor to:

store the request for information at an end of a queue; and

monitor a position of the request within the queue; and

wherein, generating the response to the request is performed in response to determining that the request is at a top of the queue;

wherein, in response to storing the request for information at the end of the queue, the instructions, when executed by the hardware processor, further cause the processor to:

collect values for a set of parameters associated with the system;

determine, based on the request for the information and the collected values for the set of parameters, an estimate of a time to generate the response to the request; and

transmit, to the user, the estimate of the time to generate the response to the request;

wherein, in response to storing the request for information at the end of the queue, the instructions, when executed by the hardware processor, further cause the processor to:

collect values for a set of parameters associated with the system;

determine, based on the request for the information and the collected values for the set of parameters, an estimate of a time to generate the response to the request; and

transmit, to the user, the estimate of the time to generate the response to the request.

10. The non-transitory computer-readable medium of claim 9 , wherein the instructions for interfacing with the first application comprise at least one of:

a custom API for interfacing with the application; and

a set of headless browser steps for interfacing with the application.

11. The non-transitory computer-readable medium of claim 9 , wherein generating the instructions for interfacing with the first application comprises executing a machine learning algorithm trained to generate the instructions for interfacing with the first application, based on a plurality of sets of instructions stored in a memory, wherein each set of instructions of the plurality of sets of instructions:

is associated with a particular process depicted in a particular video of the set of videos;

comprises instructions for interfacing with the application associated with the particular video; and

was generated based on the metadata associated with the particular process depicted in the particular video.

Assignments (1)

Continuity (2)

References Cited (56)