Utilizing machine learning with self-support actions to determine support queue positions for support calls

Patent metadata

Field	Value
Publication number	US-11882010-B2
Application number	US-202117452595-A
Country	US
Kind code	B2
Filing date	Oct 28, 2021
Priority date	Nov 5, 2018
Publication date	Jan 23, 2024
Grant date	Jan 23, 2024

Abstract

A device receives a communication associated with a support issue encountered by a user, and receives information identifying one or more self-support actions performed by the user in relation to the support issue. The device assigns the communication to a position in a support queue. The support queue includes information identifying positions of other communications received from other users, when the other communications are received, and self-support actions performed by the other users. The device associates the information identifying the one or more self-support actions with information identifying the position of the communication, and applies respective weights to the one or more self-support actions. The device generates a score for the communication based on applying the respective weights, and modifies the position of the communication based on the score. The device performs one or more actions based on modifying the position of the communication.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a first device, information identifying at least one action of a set of actions associated with self-support attempts to resolve an issue;
processing, by the first device and with a machine learning model, the information identifying the at least one action to generate respective weights for the at least one action;
generating, by the first device and based on generating the respective weights for the at least one action, a score for a communication associated with a second device;
modifying, by the first device and based on the score, a position assigned to the communication in a queue associated with the communication; and
performing, by the first device and based on modifying the position assigned to the communication in the queue, one or more other actions.

2. The method of claim 1 , wherein processing the information identifying the at least one action to generate the respective weights comprises:
processing the information identifying the at least one action and information identifying a period of time for performing the at least one action; and
generating, based on processing the information identifying the at least one action and information identifying the period of time for performing the at least one action, the respective weights.

3. The method of claim 1 , wherein a first action, of the at least one action, is associated with a first period of time,
wherein a second action, of the at least one action, is associated with a second period of time that is greater than the first period of time, and
wherein the method further comprises:
generating, based on the first period of time, a first weight, of the respective weights, for the first action; and
generating, based on the second period of time, a second weight, of the respective weights, for the second action.

4. The method of claim 1 , further comprising:
assigning, based on when the communication was received, the communication to the position in the queue.

5. The method of claim 1 , further comprising:
performing a training operation on the machine learning model using historical information associated with previously performed actions associated with previous attempts to resolve one or more issues, including the issue.

6. The method of claim 5 , wherein the historical information includes information indicating at least one of:
difficulties associated with performing the previously performed actions,
a time duration performing the previously performed actions,
results based on performing the previously performed actions, or
a time savings based on performing the previously performed actions.

7. The method of claim 1 , wherein performing the one or more other actions comprises at least one of:
providing an indication that the position assigned to the communication in the queue is modified,
providing information identifying another action, of the set of actions, that may be performed to improve the position assigned to the communication in the queue,
causing a third device to reboot,
causing the third device to execute a self-diagnostic action,
causing an autonomous vehicle to travel to a location associated with the second device,
or
causing an unmanned aerial vehicle to travel to the location.

8. A first device, comprising:
one or more memories; and
one or more processors, coupled to the one or more memories, configured to:
receive information identifying a first action associated with self-support attempts to resolve an issue;
process, with a machine learning model, the information identifying the first action to generate a respective weight associated with the first action;
generate, based on generating the respective weight, a score for a communication session associated with a second device;
modify, based on the score, a position assigned to the communication session in a queue associated with the communication session; and
perform, based on modifying the position assigned to the communication session in the queue, one or more second actions.

9. The first device of claim 8 , wherein the one or more processors, to process the information identifying the first action to generate the respective weight, are configured to:
process the information identifying the first action and information identifying a period of time for performing the first action; and
generate, based on processing the information identifying the first action and information identifying the period of time for performing the first action, the respective weight.

10. The first device of claim 8 , wherein the first action is associated with a first period of time,
wherein a third action is associated with a second period of time that is greater than the first period of time, and
wherein the one or more processors are further configured to:
generate, based on the first period of time, a first weight associated with the first action; and
generate, based on the second period of time, a second weight associated with the third action.

11. The first device of claim 8 , wherein the one or more processors are further configured to:
assign, based on when the communication session was received, the communication session to the position in the queue.

12. The first device of claim 8 , wherein the one or more processors are further configured to:
perform a training operation on the machine learning model using historical information associated with previously performed actions associated with previous attempts to resolve one or more issues, including the issue.

13. The first device of claim 12 , wherein the historical information includes information indicating at least one of:
difficulties associated with performing the previously performed actions,
a time duration performing the previously performed actions,
results based on performing the previously performed actions, or
a time savings based on performing the previously performed actions.

14. The first device of claim 8 , wherein the one or more processors, to perform the one or more second actions, are configured to:
provide an indication that the position assigned to the communication session in the queue is modified,
provide information identifying a third action that may be performed to improve the position assigned to the communication session in the queue,
cause a third device to reboot,
cause the third device to execute a self-diagnostic action,
cause an autonomous vehicle to travel to a location associated with the second device, or
cause an unmanned aerial vehicle to travel to the location.

15. A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a first device, cause the first device to:
receive information identifying at least one action, of a set of actions associated with self-support attempts to resolve an issue;
process, with a machine learning model, the information identifying the at least one action to generate respective weights for the at least one action;
generate, based on the respective weights, a score for a communication associated with a second device;
modify, based on the score, a position assigned to the communication in a queue associated with the communication; and
perform, based on modifying the position assigned to the communication in the queue, one or more other actions.

16. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the first device to process the in

Assignees

• Capital One Services Llc

Inventors

• Benkreira Abdelkadar M'Hamed
• Mossoba Michael
• Edwards Joshua

Classifications

• H04L41/5074 Primary

  Handling of user complaints or trouble tickets · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06Q30/016

  After-sales · CPC title

• H04L51/02

  using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages · CPC title