IP Library Granted Patent US 12,368,641
Granted Patent B2
US 12,368,641 · App. 18/480,537 · Granted Jul 22, 2025

Apparatuses and methods for facilitating solutions to optimization problems via modeling through natural language processing and machine learning

Inventors: Carlos Eduardo De Andrade (Gainesville, FL); Rakesh Sinha (Edison, NJ); Ajay Mahimkar (Edison, NJ)
Assignee: AT&T Intellectual Property I, L.P.
H04L41/0823G06F40/40H04L41/16
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,368,641
App. No.
18/480,537
Granted
Jul 22, 2025
Kind
B2
Abstract

Aspects of the subject disclosure may include, for example, obtaining a natural language statement of an optimization problem that is to be solved, analyzing the natural language statement to select a template included within a plurality of templates, generating a definition for the optimization problem using the template, constructing a model in accordance with the definition, resulting in a constructed model, and validating the constructed model, resulting in a validated model. Other embodiments are disclosed.

Claims (58)

1. A device, comprising:

a processing system including a processor; and

a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:

obtaining a natural language statement of an optimization problem that is to be solved;

analyzing the natural language statement to select a template included within a plurality of templates;

generating a definition for the optimization problem using the template;

constructing a model in accordance with the definition, resulting in a constructed model;

validating the constructed model, resulting in a validated model;

obtaining a user input that modifies a parameter of the validated model, resulting in a modified model; and

modifying at least one parameter of a base station of a communication network in accordance with the modified model.

2. The device of claim 1 , wherein the operations further comprise:

storing the validated model as a second template within the plurality of templates.

3. The device of claim 2 , wherein the operations further comprise:

generating metadata that represents features associated with the constructed model,

wherein the second template includes the metadata.

4. The device of claim 1 , wherein the operations further comprise:

requesting information from an operator of the device, wherein the information is lacking in the natural language statement.

5. The device of claim 4 , wherein the analyzing incorporates the information to select the template.

6. The device of claim 4 , wherein the information is used as part of generating the definition.

7. The device of claim 1 , wherein the operations further comprise:

providing the validated model to a communication device associated with a user; and

based on the providing of the validated model to the communication device associated with the user, obtaining the user input that modifies the parameter of the validated model.

8. The device of claim 7 , wherein the at least one parameter further pertains to, a router, a switch, an access point, a user equipment, or any combination thereof.

9. The device of claim 1 , wherein the validating includes performing a test against the constructed model.

10. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:

obtaining a natural language description of a problem;

based on the obtaining of the natural language description of the problem, accessing at least one template that imposes at least one constraint related to the problem;

based on the obtaining of the natural language description of the problem, selecting a model included within a plurality of models;

modifying the model in accordance with the at least one template to generate a modified model;

generating a solution to the problem based on the modified model; and

modifying at least one parameter of a router of a communication network in accordance with the modified model.

11. The non-transitory machine-readable medium of claim 10 , wherein the solution is one of a plurality of solutions to the problem.

12. The non-transitory machine-readable medium of claim 11 , wherein the solution is an optimum solution amongst the plurality of solutions in accordance with an objective function.

13. The non-transitory machine-readable medium of claim 10 , wherein the operations further comprise:

implementing the solution as part of the communication network.

14. The non-transitory machine-readable medium of claim 13 , wherein the operations further comprise:

subsequent to the implementing, generating a message or a report that provides an indication of an output of the communication network that is based on the solution.

15. The non-transitory machine-readable medium of claim 10 , wherein the operations further comprise:

requesting information from a user, wherein the information is lacking in the natural language description,

wherein the solution is based on the information.

16. The non-transitory machine-readable medium of claim 10 , wherein the obtaining of the natural language description of the problem occurs via a microphone, a keyboard, or a combination thereof.

17. The non-transitory machine-readable medium of claim 10 , wherein the at least one template is a first template included in a plurality of templates, and wherein the operations further comprise:

applying first metadata to the first template based on an analysis of first data included in the first template; and

applying second metadata to a second template of the plurality of templates based on an analysis of second data included in the second template,

wherein the second data is different from the first data and the second metadata is different from the first metadata.

18. A method, comprising:

obtaining, by a processing system including a processor, a natural language statement of an optimization problem pertaining to a communication network;

generating, by the processing system, a definition of the optimization problem based on an analysis of the natural language statement;

determining, by the processing system and based on the definition, that information or data is missing from the natural language statement;

requesting, by the processing system, the information or data from an operator;

obtaining, by the processing system and based on the requesting, the information or data from the operator, resulting in obtained information or data;

constructing, by the processing system, a model based on the obtained information or data, the model providing at least one solution to the optimization problem; and

modifying at least one parameter of a switch of the communication network in accordance with the model.

19. The method of claim 18 , wherein the obtaining of the natural language statement comprises obtaining the natural language statement from the operator.

20. The method of claim 18 , wherein the at least one solution comprises a plurality of solutions, the method further comprising:

presenting, by the processing system, indications of the plurality of solutions to the operator;

obtaining, by the processing system and based on the presenting, a selection of a solution included in the plurality of solutions; and

implementing, by the processing system, the solution.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 4, 2023
From: DE ANDRADE, CARLOS EDUARDO; SINHA, RAKESH; MAHIMKAR, AJAY
To: AT&T INTELLECTUAL PROPERTY I, L.P.
Reel/Frame 065119/0475 →
Continuity (1)
Related Publication 20250119350A1 · Apr 10, 2025
References Cited (8)
US 10958517B2 · Mahimkar et al. · 2021 [cited by applicant]
US 11106569B1 · Chapagain · 2021 [cited by examiner]
US 11463307B2 · Mahimkar et al. · 2022 [cited by applicant]
US 20170293871A1 · Ceugniet · 2017 [cited by examiner]
US 20210018881A1 · Rezgui · 2021 [cited by examiner]
US 20220107786A1 · Iwasaki · 2022 [cited by examiner]
US 20220321408A1 · Mahimkar et al. · 2022 [cited by applicant]
US 20230108774A1 · Archer · 2023 [cited by examiner]