Virtual agent conversation service

Patent metadata

Field	Value
Publication number	US-10673787-B2
Application number	US-201715723795-A
Country	US
Kind code	B2
Filing date	Oct 3, 2017
Priority date	Oct 3, 2017
Publication date	Jun 2, 2020
Grant date	Jun 2, 2020

Abstract

Techniques related to a virtual agent conversation service are disclosed. The techniques include a method for improving performance of a virtual agent conversation system, comprising receiving one or more conversation trees, receiving a message from a user including a natural language description of an intended task, providing the message to a conversation provider, receiving an indication of an intent of the message and textual information related to the indication of the intent, determining an action to take based on the indication of the intent, perform the action to obtain an action result, determining a set of responses based on the action result, and providing a response of the set of responses to the virtual agent based on the action result for output to the user.

First claim

What is claimed is:

1. A system, comprising:
one or more non-transitory memory devices; and
one or more hardware processors configured to execute instructions from the one or more non-transitory memory devices to cause the system to:
generate a conversation tree based on one or more forms, wherein the one or more forms comprise a catalog item or a survey, wherein the conversation tree is associated with at least one or more intents and one or more actions, wherein generating the conversation tree comprises:
determining a set of catalog items available, wherein the set of catalog items comprises the catalog item;
generating a set of entities based on a set of words associated with the catalog item;
determining one or more variables associated with the catalog item; and
generating the conversation tree based on the catalog item, the set of entities, and the one or more variables;

provide, to a conversation provider, the conversation tree;
receive, via a virtual agent, a message from a user including a natural language description of an intended task;
provide the message to the conversation provider;
receive, from the conversation provider, an indication of an intent of the message and textual information related to the indication of the intent, wherein the intent is from the at least one or more intents;
invoke the conversation tree to determine an action to take based on the indication of the intent; perform the action to obtain an action result;
determine a set of responses based on the action result; and
provide a response of the set of responses to the virtual agent based on the action result for output to the user.

2. The system of claim 1 , wherein the indication of the intent comprises a parameter and wherein the action comprises running a script associated with the action.

3. The system of claim 1 , wherein the instructions are further configured to cause the system to:
determine validation rules based on a set of behavioral rules associated with the one or more forms and wherein the provided response is selected from the set of responses;
receive a user response to the provided response;
determine the user response is a valid response based on the validation rules; and
associate the user response with a corresponding field of the one or more forms.

4. The system of claim 1 , wherein the message is received as an email message and the response is provided as a reply to the email message.

5. The system of claim 1 , wherein the instructions are further configured to cause the system to:
receive a user notification for outputting to the user;
determine a notification channel for communicating the user notification, wherein the virtual agent supports the notification channel; and
transmit, by the virtual agent, the user notification using the determined notification channel.

6. The system of claim 5 , wherein the notification channel comprises multiple notification channels and transmitting the user notification comprises transmitting the user notification on the multiple notification channels.

7. The system of claim 1 , wherein the response comprises a confirmation question for the intent based on a field description on the one or more forms.

8. The system of claim 1 , wherein the instructions are further configured to cause the system to:
generate a plurality of conversation trees, including the conversation tree, based on the one or more forms; and
identify the conversation tree from the plurality of conversation trees, as associated with the intent.

9. The system of claim 1 , wherein generating the conversation tree comprises:
parsing a plurality of catalog items based on variables associated with each catalog item of the plurality of catalog items, wherein each catalog item is associated with a corresponding form built by the associated variables;
generating a plurality of conversation trees including the conversation tree, wherein each conversation tree of the plurality of conversation trees is associated with a corresponding catalog item.

10. A method for improving a virtual agent conversation service, comprising:
generating a conversation tree based on one or more forms, wherein the one or more forms comprise a catalog item or a survey, wherein the conversation tree is associated with at least one or more intents and one or more actions, wherein generating the conversation tree comprises:
determining a set of catalog items available, wherein the set of catalog items comprises the catalog item;
generating a set of entities based on a set of words associated with the catalog item of the set of catalog items;
determining one or more variables associated with the catalog item; and
generating the conversation tree based on the catalog item, the set of entities, and the one or more variables;

providing, to a conversation provider, the conversation tree;
receiving, via a virtual agent, a message from a user including a natural language description of an intended task;
providing the message to the conversation provider;
receiving, from the conversation provider, an indication of an intent of the message and textual information related to the indication of the intent, wherein the intent is from the at least one or more intents;
invoking the conversation tree to determine an action to take based on the indication of the intent;
performing the action to obtain an action result;
determining a set of responses based on the action result; and
providing a response of the set of responses to the virtual agent based on the action result for output to the user.

11. The method of claim 10 , wherein the indication of the intent comprises a parameter and wherein the action comprises running a script associated with the action.

12. The method of claim 10 , further comprising:
determining validation rules based on a set of behavioral rules associated with a form of the one or more forms and wherein the provided response is selected from the set of responses;
receiving a user response to the provided response;
determining the user response is a valid response based on the validation rules; and
associating the user response with a corresponding field of the form.

13. The method of claim 10 , wherein the message is received as an email message and the response is provided as a reply to the email message.

14. The method of claim 10 , further comprising:
receiving a user notification for outputting to the user;
determining a notification channel for communicating the user notification, wherein the virtual agent supports the notification channel; and
transmitting, by the virtual agent, the user notification using the determined notification channel.

15. The method of claim 14 , wherein the notification channel comprises multiple notification channels and transmitting the user notification comprises transmitting the user notification on the multiple notification channels.

16. The method of claim 10 , wherein the response comprises a confirmation question for the intent based on a field description on the form.

17. A non-transitory computer readable medium containing instructions that, when executed by a processor cause a programmable device:
generate a conversation tree based on one or more forms, wherein the one or more forms comprise a catalog item or a survey, wherein the conversation tree is associated with at least one or more intents and one or more actions, wherein generating the conversation tree comprises:
determining a set of catalog items available, wherein the set of catalog items comprises the catalog item;
gen

Assignees

• Servicenow Inc

Inventors

• Koohmarey Darius
• Arobone Anthony
• Mendoza John Oliver
• Venkatachalam Gopalakrishnan
• Holm Nathan
• Doshi Naitik

Classifications

• G06Q30/0641

  utilising user interfaces specially adapted for shopping · CPC title

• H04L51/02 Primary

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

• H04L51/18

  Commands or executable codes · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06Q30/0203 Primary

  Market surveys; Market polls · CPC title