IP Library › Granted Patent US 11,425,215
Granted Patent B1
US 11,425,215 · App. 16/111,808 · Granted Aug 23, 2022

Methods and systems for virtual assistant routing

Inventors: Olvin Brett Lewis (New Braunfels, TX); Justin Leggett (Bulverde, TX); Guy R. Langley (San Antonio, TX); Andrew Jamison (San Antonio, TX)
Assignee: United Services Automobile Association (USAA)
H04L67/327G06F9/4881G06F16/90332G06F40/30G10L15/22H04M3/527G10L2015/223
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 11,425,215
App. No.
16/111,808
Granted
Aug 23, 2022
Kind
B1
Abstract

Disclosed herein are embodiments of systems, methods, and products comprises a server, which receive a request from a user's electronic client device. The server understands the intention of the user by determining an attribute associated with the request. The server further determines a sentiment value and routes the request to a call center computing system if the sentiment value satisfies a threshold; otherwise, the server determines to route the request to one of virtual assistant servers. The server selects a virtual assistant server whose subject matter best matches the attribute of the request. The server may also select a virtual assistant server based on other information, such as user's previous selection of virtual assistant servers, a time value, or a confidence value. The selected virtual assistant provides a response corresponding to the request.

Claims (63)

1. A computer-implemented method comprising:

receiving, by a server, a first request from an electronic client device, wherein the first request is inputted by a user into a virtual assistant application executing on the electronic client device;

executing, by the server, a natural language processing protocol to determine a first attribute and a first sentiment value associated with the first request, the first attribute corresponding to a subject matter of the first request, and the first sentiment value corresponding to one or more phrases within the first request;

determining a sentiment threshold, based on the user and the subject matter, that specifies a sentiment level for differentiating between call center or virtual assistant destinations for one or more requests;

when the first sentiment value satisfies the sentiment threshold, routing, by the server, the first request to a call center computing system;

when the first sentiment value does not satisfy the sentiment threshold, routing, by the server, the first request to a first virtual assistant server from a plurality of virtual assistant servers,

wherein the first virtual assistant server is selected based at least on one of the first attribute or the first sentiment value, and

wherein each virtual assistant server is configured to at least partially satisfy requests;

updating a user profile to indicate the selected first virtual assistant server;

upon receiving a second request from the electronic client device:

determining, by the server, external context metadata of the user, wherein the external context metadata of the user comprises information outside the second request;

executing, by the server, the natural language processing protocol to determine a second attribute and a second sentiment value associated with the second request; and

routing, by the server, the second request to a second virtual assistant server from the plurality of virtual assistant servers or the call center computing system based on the external context metadata of the user, the second attribute and the second sentiment value,

wherein the server further transmits the external context metadata of the user, metadata corresponding to the first request comprising:

at least one of the first attribute or the first sentiment value,

the first request, a response to the first request, and

the updated user profile to the second virtual assistant server or the call center computing system.

2. The method of claim 1 , wherein the requests are received in response to the user's auditory input.

3. The method of claim 1 , wherein the selection of the virtual assistant server comprises a round robin selection.

4. The method of claim 1 , wherein the selection of the virtual assistant server comprises a random selection.

5. The method of claim 1 , wherein the selection of the virtual assistant server comprises a shotgun method.

6. The method of claim 1 , wherein the selection of the virtual assistant server comprises a serial method.

7. The method of claim 1 , further comprising:

selecting, by the server, a virtual assistant server based on a previous selection of virtual assistant servers associated with the user by utilizing machine learning algorithms.

8. The method of claim 1 , further comprising:

transforming, by the server, the received request from the electronic client device into computer readable type for the server; and

transforming, by the server, the response from the selected virtual assistant server into client device readable type for the electronic client device.

9. The method of claim 1 , wherein the server selects the selected virtual assistant server based on at least one of a shortest time value, a highest confidence value, and a highest similarity value.

10. The method of claim 1 , further comprising:

adjusting the sentiment threshold for the user, wherein the sentiment threshold is configured based on historical sentiment values for the user.

11. A system comprising:

an electronic client device;

a plurality of virtual assistant servers;

a call center computing device;

a server in communication with the plurality of virtual assistant servers and the call center computing device and configured to:

receive a first request from the electronic client device, wherein the first request is inputted by a user into a virtual assistant application executing on the electronic client device;

execute a natural language processing protocol to determine a first attribute and a first sentiment value associated with the first request, the first attribute corresponding to a subject matter of the first request, and the first sentiment value corresponding to one or more phrases within the first request;

determine a sentiment threshold, based on the user and the subject matter, that specifies a sentiment level for differentiating between call center or virtual assistant destinations for one or more requests;

when the first sentiment value satisfies the sentiment threshold, route the first request to the call center computing system;

when the first sentiment value does not satisfy the sentiment threshold, route the first request to a first virtual assistant server from the plurality of virtual assistant servers,

wherein the first virtual assistant server is selected based at least on one of the first attribute or the first sentiment value, and

wherein each virtual assistant server is configured to at least partially satisfy requests;

updating a user profile to indicate the selected first virtual assistant server;

upon receiving a second request from the electronic client device:

determine, by the server, external context metadata of the user, wherein the external context metadata of the user comprises information outside the second request;

execute the natural language processing protocol to determine a second attribute and a second sentiment value associated with the second request; and

route the second request to a second virtual assistant server from the plurality of virtual assistant servers or the call center computing device based on the external context metadata of the user, the second attribute and the second sentiment value,

wherein the server further transmits the external context metadata of the user, metadata corresponding to the first request comprising:

 at least one of the first attribute or the first sentiment value, the first request,

 a response to the first request, and

 the updated user profile to the second virtual assistant server or to the call center computing device.

12. The system of claim 11 , wherein the requests are received in response to the user's auditory input.

13. The system of claim 11 , wherein the selection of the virtual assistant server comprises a round robin selection.

14. The system of claim 11 , wherein the selection of the virtual assistant server comprises a random selection.

15. The system of claim 11 , wherein the selection of the virtual assistant server comprises a shotgun method.

16. The system of claim 11 , wherein the selection of the virtual assistant server comprises a serial method.

17. The system of claim 11 , wherein the server is further configured to:

select a virtual assistant server based on a previous selection of virtual assistant servers associated with the user by utilizing machine learning algorithms.

18. The system of claim 11 , wherein the server is further configured to:

transform the received request from the electronic client device into computer readable type for the server; and

transform the response from the selected virtual assistant server into client device readable type for the electronic client device.

19. The system of claim 11 , wherein the server selects the selected virtual assistant server based on at least one of a shortest time value, a highest confidence value, and a highest similarity value.

20. The system of claim 11 , wherein the server is further configured to adjust the sentiment threshold for the user, wherein the sentiment threshold is configured based on historical sentiment values for the user.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 14, 2022
From: UIPCO, LLC
To: UNITED SERVICES AUTOMOBILE ASSOCIATION (USAA)
Reel/Frame 060510/0692 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 24, 2018
From: LEWIS, OLVIN BRETT; LEGGETT, JUSTIN; LANGLEY, GUY R.; JAMISON, ANDREW P.
To: UIPCO, LLC.
Reel/Frame 046697/0172 →
Continuity (1)
Provisional Application 62549875 · Aug 24, 2017
Cited By (5)
US 12,233,797 US 12,278,925 US 12,321,840 US 12,688,209 US 12,744,848