Chat bot operation and management architecture

Patent metadata

Field	Value
Publication number	US-2020412671-A1
Application number	US-201916535273-A
Country	US
Kind code	A1
Filing date	Aug 8, 2019
Priority date	Jun 27, 2019
Publication date	Dec 31, 2020
Grant date	—

Abstract

An improved chat bot operation enables multiple teams to leverage a common bot deployment, rather than requiring each team to build and deploy their own. A context-aware operation identifies a user's context and selects a context file, from among a plurality of context files, to tailor actions and responses. Each team thus has a reduced workload in generating a context file rather than an entire bot deployment. An exemplary method includes: receiving a first chat content from a first chat session; determining a first context for the first chat content; selecting, based at least on the first context, a first context file from a plurality of context files; determining, based at least on the first chat content and the first context file, a first action for the chat bot, wherein determining the first action for the chat bot comprises parsing the first context file; and executing the first action.

First claim

What is claimed is:

1 . A system for managing operations of a chat bot, the system comprising:
a processor; and a computer-readable medium storing instructions that are operative when executed by the processor to:
receive a first chat content from a first chat session;
determine a first context for the first chat content;
select, based at least on the first context, a first context file from a plurality of context files;
determine, based at least on the first chat content and the first context file, a first action for the chat bot, wherein determining the first action for the chat bot comprises parsing the first context file; and
execute the first action.

2 . The system of claim 1 wherein the first context corresponds with a first organization of a plurality of organizations using the chat bot.

3 . The system of claim 1 wherein the first context file comprises a function for accessing an external resource, and wherein executing the action comprises:
executing the function; and
reporting a result of executing the function to the first chat session.

4 . The system of claim 1 wherein the instructions are further operative to:
receive a second chat content from a second chat session, different than the first chat session;
determine a second context for the second chat content, the second context different than the first context;
select, based at least on the second context, a second context file from the plurality of context files, the second context file different than the first context file;
determine, based at least on the second chat content and the second context file, a second action for the chat bot, wherein determining the second action for the chat bot comprises parsing the second context file; and
execute the second action.

5 . The system of claim 4 wherein the instructions are further operative to:
generate, based at least on the second context, the second context file;
receive the second context file; and
add the second context file to the plurality of context files.

6 . The system of claim 1 wherein the plurality of context files comprises a plurality of YAML files.

7 . The system of claim 1 further comprising:
a data acquisition sensor; and
wherein executing the first action comprises:
controlling the data acquisition sensor to collect data; and
reporting the collected data to the first chat session.

8 . The system of claim 1 wherein determining a first context for the first chat content comprises determining, using artificial intelligence (AI), a first context for the first chat content.

9 . The system of claim 1 wherein the instructions are further operative to:
enhance the plurality of context files with additional functionality using a machine learning (ML) component.

10 . A method of managing operations of a chat bot, the method comprising:
receiving a first chat content from a first chat session; determining a first context for the first chat content; selecting, based at least on the first context, a first context file from a plurality of context files; determining, based at least on the first chat content and the first context file, a first action for the chat bot, wherein determining the first action for the chat bot comprises parsing the first context file; and executing the first action.

11 . The method of claim 10 wherein the first context corresponds with a first organization of a plurality of organizations using the chat bot.

12 . The method of claim 10 wherein the first context file comprises a function for accessing an external resource, and wherein executing the action comprises:
executing the function; and
reporting a result of executing the function to the first chat session.

13 . The method of claim 10 further comprising:
receiving a second chat content from a second chat session, different than the first chat session;
determining a second context for the second chat content, the second context different than the first context;
selecting, based at least on the second context, a second context file from the plurality of context files, the second context file different than the first context file;
determining, based at least on the second chat content and the second context file, a second action for the chat bot, wherein determining the second action for the chat bot comprises parsing the second context file; and
executing the second action.

14 . The method of claim 13 further comprising:
generating, based at least on the second context, the second context file;
receiving the second context file; and
adding the second context file to the plurality of context files.

15 . The method of claim 10 wherein the plurality of context files comprises a plurality of YAML files.

16 . The method of claim 10 wherein executing the first action comprises:
controlling a data acquisition sensor to collect data; and
reporting the collected data to the first chat session.

17 . The method of claim 10 wherein determining a first context for the first chat content comprises determining, using artificial intelligence (AI), a first context for the first chat content.

18 . The method of claim 10 further comprising:
enhancing the plurality of context files with additional functionality using a machine learning (ML) component.

19 . One or more computer storage devices having computer-executable instructions stored thereon for managing operations of a chat bot, which, on execution by a computer, cause the computer to perform operations comprising:
receiving a first chat content from a first chat session; determining a first context for the first chat content; selecting, based at least on the first context, a first context file from a plurality of context files, wherein the first context file comprises a function for accessing an external resource; determining, based at least on the first chat content and the first context file, a first action for the chat bot, wherein determining the first action for the chat bot comprises parsing the first context file; executing the first action, wherein executing the first action comprises:
executing the function; and
reporting a result of executing the function to the first chat session;
generating, based at least on a second context, a second context file; receiving the second context file; adding the second context file to the plurality of context files; receiving a second chat content from a second chat session, different than the first chat session; determining the second context for the second chat content, the second context different than the first context; selecting, based at least on the second context, the second context file from the plurality of context files, the second context file different than the first context file; determining, based at least on the second chat content and the second context file, a second action for the chat bot, wherein determining the second action for the chat bot comprises parsing the second context file; executing the second action, wherein executing the second action comprises:
controlling a data acquisition sensor to collect data; and
reporting the collected data to the second chat session.

20 . The one or more computer storage devices of claim 19 wherein determining a first context for the first chat content comprises determining, using artificial intelligence (AI), a first context for the first chat content, and wherein the operations further comprise:
enhancing the plurality of

Assignees

• Walmart Apollo Llc

Inventors

• Mathur Yadu
• Chetluru Revanth

Classifications

• G06N20/00

  Machine learning · CPC title

• H04L65/1069

  Session establishment or de-establishment · CPC title

• H04L65/4015

  where at least one of the additional parallel sessions is real time or time sensitive, e.g. white board sharing, collaboration or spawning of a subconference · CPC title

• G06F40/35

  Discourse or dialogue representation · CPC title

• G06F40/205

  Parsing · CPC title