TECHNIQUES FOR AUTOMATED DETERMINATION OF FORM RESPONSES

• US 20160285816A1
• Filed: 07/21/2015
• Published: 09/29/2016
• Est. Priority Date: 03/25/2015
• Status: Abandoned Application

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Schmid, Matthew Logan, Bush, Matthew Sanchez, Vulkan, Roee, Sharon, Eyal Michael
• Application Number:
  US14/804,991
• Publication Number:
  US 20160285816A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0635 Processing of requisition or of purchase orders
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/52 for supporting social networking services
• CPC Class Codes: H04L 67/306 User profiles

Abstract

Techniques for automated determination of form responses are described. In one embodiment, a commerce intermediary server may be operative to receive a messaging package from a messaging endpoint at a messaging service; perform a natural language processing of the messaging package to determine a response message predicted to correspond to the messaging package; determine an administrator account assigned to respond to the messaging package; and transmit a form response package to an administrator messaging endpoint for the administrator account, the form response package comprising the response message. Other embodiments are described and claimed.

Claims

1. A computer-implemented method, comprising:

receiving a messaging package from a messaging endpoint at a messaging service;

performing a natural language processing of the messaging package to determine a response message predicted to correspond to the messaging package;

determining an administrator account assigned to respond to the messaging package; and

transmitting a form response package to an administrator messaging endpoint for the administrator account, the form response package comprising the response message.

2. The method of claim 1, the form response package operative to configure the administrator messaging endpoint to respond to the messaging package via the messaging service with the response message.

3. The method of claim 1, the response message comprising one or more dynamic fields, further comprising:

retrieving user account information associated with the messaging endpoint; and

assigning one or more values to at least one of the one or more dynamic fields based on the user account information.

4. The method of claim 1, wherein performing the natural language processing comprises submitting at least a portion of the messaging package to a natural language processor.

5. The method of claim 4, the natural language processor trained based on a messaging interaction history for the messaging service.

6. The method of claim 4, further comprising:

receiving one or more messaging intent keywords from the natural language processor; and

retrieving the response message from a response message repository based on the one or more messaging intent keywords.

7. The method of claim 1, further comprising:

submitting a plurality of response messages to a natural language processor;

receiving an intent keyword mapping from the natural language processor in response to submitting the plurality of response messages to the natural language processor, the intent keyword mapping defining associations between the plurality of response messages and a plurality of intent keywords; and

storing the defined associations between the plurality of response messages and the plurality of intent keywords in a response message repository.

8. The method of claim 7, further comprising:

receiving a responding messaging package from the administrator messaging endpoint;

determining that the responding messaging package is not based on the response message;

extracting an administrator-specified response message from the responding messaging package;

extracting a user message from the messaging package;

submitting the user message and the administrator-specified response message to the natural language processor;

receiving an additional keyword mapping from the natural language processor, the additional keyword mapping defining an association between the administrator-specified response message and one or more additional intent keywords;

storing the administrator-specified response message in the response message repository; and

storing the association between the administrator-specified response message and the one or more additional intent keywords in the response message repository.

9. The method of claim 1, further comprising:

receiving a responding messaging package from the administrator messaging endpoint;

determining whether the responding messaging package uses the response message; and

training the natural language processor based on whether the responding messaging package uses the response message.

10. The method of claim 1, the natural language processing of the messaging package determining one or more intent keywords and one or more parameters, further comprising:

configuring an automated commerce interaction based on the one or more intent keywords and the one or more parameters.

11. An apparatus, comprising:

a processor circuit on a device;

a network interface;

a commerce intermediary server operative on the processor circuit to receive, from the network interface, a messaging package from a messaging endpoint at a messaging service;

perform a natural language processing of the messaging package to determine a response message predicted to correspond to the messaging package;

determine an administrator account assigned to respond to the messaging package; and

transmit a form response package to an administrator messaging endpoint for the administrator account, the form response package comprising the response message, the form response package operative to configure the administrator messaging endpoint to respond to the messaging package via the messaging service with the response message.

12. The apparatus of claim 11, further comprising:

the commerce intermediary server operative to submit a plurality of response messages to a natural language processor;

receive an intent keyword mapping from the natural language processor in response to submitting the plurality of response messages to the natural language processor, the intent keyword mapping defining associations between the plurality of response messages and a plurality of intent keywords;

store the defined associations between the plurality of response messages and the plurality of intent keywords in a response message repository;

receive one or more messaging intent keywords from the natural language processor; and

retrieve the response message from the response message repository based on the one or more messaging intent keywords.

13. The apparatus of claim 11, further comprising:

the commerce intermediary server operative to receive a responding messaging package from the administrator messaging endpoint;

determine that the responding messaging package is not based on the response message;

extract an administrator-specified response message from the responding messaging package;

extract a user message from the messaging package;

submit the user message and the administrator-specified response message to the natural language processor;

receive an additional keyword mapping from the natural language processor, the additional keyword mapping defining an association between the administrator-specified response message and one or more additional intent keywords;

store the administrator-specified response message in the response message repository; and

store the association between the administrator-specified response message and the one or more additional intent keywords in the response message repository.

14. The apparatus of claim 11, further comprising:

the commerce intermediary server operative to receive a responding messaging package from the administrator messaging endpoint;

determine whether the responding messaging package uses the response message; and

train the natural language processor based on whether the responding messaging package uses the response message.

15. The apparatus of claim 11, the response message comprising one or more dynamic fields, further comprising:

the commerce intermediary server operative to retrieve user account information associated with the messaging endpoint from a social networking service; and

assign one or more values to at least one of the one or more dynamic fields based on the user account information.

16. At least one computer-readable storage medium comprising instructions that, when executed, cause a system to:

receive a messaging package from a messaging endpoint at a messaging service;

perform a natural language processing of the messaging package to determine a response message predicted to correspond to the messaging package;

determine an administrator account assigned to respond to the messaging package; and

transmit a form response package to an administrator messaging endpoint for the administrator account, the form response package comprising the response message, the form response package operative to configure the administrator messaging endpoint to respond to the messaging package via the messaging service with the response message.

17. The computer-readable storage medium of claim 16, comprising further instructions that, when executed, cause a system to:

submit a plurality of response messages to a natural language processor;

receive an intent keyword mapping from the natural language processor in response to submitting the plurality of response messages to the natural language processor, the intent keyword mapping defining associations between the plurality of response messages and a plurality of intent keywords;

store the defined associations between the plurality of response messages and the plurality of intent keywords in a response message repository;

receive one or more messaging intent keywords from the natural language processor; and

retrieve the response message from the response message repository based on the one or more messaging intent keywords.

18. The computer-readable storage medium of claim 16, comprising further instructions that, when executed, cause a system to:

receive a responding messaging package from the administrator messaging endpoint;

determine that the responding messaging package is not based on the response message;

extract an administrator-specified response message from the responding messaging package;

extract a user message from the messaging package;

submit the user message and the administrator-specified response message to the natural language processor;

receive an additional keyword mapping from the natural language processor, the additional keyword mapping defining an association between the administrator-specified response message and one or more additional intent keywords;

store the administrator-specified response message in the response message repository; and

store the association between the administrator-specified response message and the one or more additional intent keywords in the response message repository.

19. The computer-readable storage medium of claim 16, comprising further instructions that, when executed, cause a system to:

receive a responding messaging package from the administrator messaging endpoint;

determine whether the responding messaging package uses the response message; and

train the natural language processor based on whether the responding messaging package uses the response message.

20. The computer-readable storage medium of claim 16, the response message comprising one or more dynamic fields, comprising further instructions that, when executed, cause a system to:

retrieve user account information associated with the messaging endpoint from a social networking service; and

assign one or more values to at least one of the one or more dynamic fields based on the user account information.