Automating communications using an intent classifier

• US 10,497,004 B2
• Filed: 11/13/2018
• Issued: 12/03/2019
• Est. Priority Date: 12/08/2017
• Status: Active Grant

• Current Assignee: Asapp, Inc.
• Original Assignee: Asapp, Inc.
• Inventors: Shaev, Jason, Sehrawat, Vicky, Knaster, Rachel, Wang, Shang Wei, Sapoznik, Gustavo
• Primary Examiner(s): McCormick, Gabrielle A
• Application Number:
  US16/189,536
• Publication Number:
  US 20190180288A1
• Time in Patent Office: 385 Days
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G06F 40/56 Natural language generation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06Q 30/016 After-sales
• CPC Class Codes: G06Q 30/0281 Customer communication at a business location, e.g. providing product or service information, consulting
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/214 using selective forwarding
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 67/148 Migration or transfer of sessions
• CPC Class Codes: H04M 3/51 Centralised call answering arrangements requiring operator intervention , e.g. call or contact centers for telemarketing
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends
• CPC Class Codes: H04M 3/5232 Call distribution algorithms

Abstract

An intent classifier may be used to increase the efficiency of a communications system. A company may provide assistance to a first user using automated processing or by a second user manually responding to the first user. To reduce costs, the company may prefer to use automated processing for assistance where it is available. While a second user is assisting a user, a message from the first user may be processed with an intent classifier to determine that automated support is available to assist the first user, and a suggestion may be presented to the second user to transfer the first user to the automated processing, such as by presenting a button to the second user to transfer the first user to the automated processing. The second user may then transfer the first user to the automated processing and assist other users.

Claims

1. A computer-implemented method for using an intent classifier for automating communications, the method comprising:

establishing a communications session with a first user, wherein the communications session includes a first network connection between a first device of the first user and a first server;

receiving, at the first server via the first network connection, first text from the first user requesting assistance;

transmitting, by the first server, the first text to a user interface of a second device of a second user, wherein the second user is assisting the first user and wherein the first text is transmitted via a second network connection between a second server and the second device of the second user;

processing the first text with the intent classifier to compute a score for each intent of a plurality of possible intents;

selecting at least two intents from the plurality of possible intents using the scores, wherein the at least two intents relate to the assistance requested by the first user;

determining, using the at least two intents, that a plurality of automated processing tasks is available to assist the first user with the assistance requested by the first user;

transmitting a listing of the plurality of automated processing tasks to the second user via the user interface;

receiving a communication from the second user via the user interface to transfer the first user the automated processing by selecting an automated processing task from the plurality of automated processing tasks;

assisting the first user with the automated processing, wherein the automated processing comprises;

receiving, at the first server via the first network connection, a communication from the first device of the first user,

determining a response to the communication from the first device using on automated processing server, and

transmitting, by the first server via the first network connection, the response to the first device of the first user.

2. The computer-implemented method of claim 1, wherein:

transmitting the listing of the plurality of automated processing tasks to the second user via the user interface comprises causing a user interface element to be presented to the second user; and

the communication from the second user via the user interface to transfer the first user to the automated processing corresponds to a selection of the user interface element.

3. The computer-implemented method of claim 1, wherein the assistance requested by the first user corresponds to updating an address of the first user, processing a payment, changing a PIN of the first user, or changing services received by the first user.

4. The computer-implemented method of claim 1, wherein the first network connection is a persistent connection, and wherein the persistent connection is used for receiving the first text and assisting the first user using the automated processing.

5. The computer-implemented method of claim 4, wherein the persistent connection comprises a websocket or long polling.

6. The computer-implemented method of claim 1, wherein:

transmitting the first text to the second user comprises transmitting the first text via a connecting server;

the connecting server allows other servers to publish to channels and to subscribe to the channels;

the first server publishes the first text to a channel of the connecting server; and

the first server subscribes to the channel of the connecting server.

7. The computer-implemented method of claim 1, wherein processing the first text with the intent classifier comprises obtaining word embeddings for words of the first text.

8. The computer-implemented method of claim 1, wherein selecting the at least two intents comprises using information about services received by the first user, a location of the first user, or a status of services provided by a company.

9. A system for providing customer support, the system comprising:

at least one server comprising at least one processor and at least one memory, the at least one server configured to;

establish a communications session with a first user, wherein the communications session includes a first network connection between a first device of the first user and a first server;

receive at the first server via the first network connection, first text from the first user requesting assistance;

transmit, by the first server, the first text to a user interface of a second device of a second user, wherein the second user is assisting the first user and wherein the first text is transmitted via a second network connection between a second server and the second device of the second user;

process the first text with an intent classifier to compute a score for each intent of a plurality of possible intents;

select at least two intents from the plurality of possible intents using the scores, wherein the at least two intents relate to the assistance requested by the first user;

determine, using the at least two intents, that a plurality of automated processing tasks is available to assist the first user with the assistance requested by the first user;

transmit a listing of the plurality of automated processing tasks to the second user via the user interface;

receive a communication from the second user via the user interface to transfer the first user to the automated processing by selecting an automated processing task from the plurality of automated processing tasks;

assist the first user with the automated processing using the first network connection.

10. The system of claim 9, wherein the at least one server is further configured to:

receive, at the first server, second text from the first user requesting second assistance;

transmit the second text to the user interface;

process the second text with the intent classifier to compute a second score for each intent of the plurality of possible intents;

select at least one second intent from the plurality of possible intents using the scores, wherein the second intent relates to the second assistance requested by the first user; and

determine, using the at least one second intent, that automated processing is not available to assist the first user with the second assistance requested by the first user.

11. The system of claim 9, wherein the first network connection is a persistent connection, and wherein the persistent connection is used for receiving the first text and assisting the first user using the automated processing.

12. The system of claim 9, wherein the automated processing comprises transmitting an API request to a server of a company providing a service to the first user.

13. The system of claim 9, wherein the at least one server is further configured to:

update a database entry to indicate that the first user is receiving support from the second user; and

update the database entry to indicate that the first user is receiving support using the automated processing after receiving the communication from the second user to assist the first user using the automated processing.

14. The system of claim 9, wherein the at least one server is configured to:

transmit the first text to the second user by transmitting the first text via a connecting server;

the connecting server allows other servers to publish to channels and to subscribe to the channels;

the first server publishes the first text to a channel of the connecting server; and

the first server subscribes to the channel of the connecting server.

15. The system of claim 9, wherein the at least one server is operated by a third-party company that provides services to a company and the first user is a customer of the company.

16. One or more non-transitory, computer-readable media comprising computer executable instructions that, when executed, cause at least one processor to perform actions comprising:

establishing a communications session with a first user, wherein the communications session includes a first network connection between a first device of the first user and a first server;

receiving, at the first server via the first network connection, first text from the first user requesting assistance;

transmitting, by the first server, the first text to a user interface of a second device of a second user, wherein the second user is assisting the first user and wherein the first text is transmitted via a second network connection between a second server and the second device of the second user;

processing the first text with an intent classifier to compute a score for each intent of a plurality of possible intents;

selecting at least two intents from the plurality of possible intents using the scores, wherein the at least two intents relate to the assistance requested by the first user;

determining, using the at least two intents, that a plurality of automated processing tasks is available to assist the first user with the assistance requested by the first user;

transmitting a listing of the plurality of automated processing tasks to the second user via the user interface;

receiving a communication from the second user via the user interface to transfer the first user to the automated processing by selecting an automated processing task from the plurality of automated processing tasks; and

assisting the first user with the automated processing using the first network connection.

17. The one or more non-transitory, computer-readable media of claim 16, wherein selecting the at least two intents comprises using information about services received by the first user, a location of the first user, or a status of services provided by a company.

18. The one or more non-transitory, computer-readable media of claim 16, wherein the first network connection from the first server to the first device is a persistent connection, and wherein the persistent connection is used for receiving the first text and assisting the first user using the automated processing.

19. The one or more non-transitory, computer-readable media of claim 16, wherein the intent classifier comprises a neural network.

20. The one or more non-transitory, computer-readable media of claim 16, wherein assisting the first user using the automated processing comprises transmitting a text message to the first user or presenting a form field to the first user.