System for focused conversation context management in a reasoning agent/behavior engine of an agent automation system

Patent metadata

Field	Value
Publication number	US-11087090-B2
Application number	US-201916239147-A
Country	US
Kind code	B2
Filing date	Jan 3, 2019
Priority date	Mar 23, 2018
Publication date	Aug 10, 2021
Grant date	Aug 10, 2021

Abstract

An agent automation system includes a memory configured to store a reasoning agent/behavior engine (RA/BE) including a first persona and a current context and a processor configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising: receiving intents/entities of a first user utterance; recognizing a context overlay cue in the intents/entities of the first user utterance, wherein the context overlay cue defines a time period; updating the current context of the RA/BE by overlaying context information from at least one stored episode associated with the time period; and performing at least one action based on the intents/entities of the first user utterance and the current context of the RA/BE.

First claim

What is claimed is:

1. An agent automation system, comprising:
a memory configured to store a reasoning agent/behavior engine (RA/BE) comprising a first persona, a current context, and a demultiplexer; and
a processor configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
receiving intents/entities of a first user utterance, wherein the demultiplexer is configured to provide the intents/entities of the first user utterance to the first persona of the RA/BE, or another persona of the RA/BE, based on one or more stored rules;
recognizing a context overlay cue in the intents/entities of the first user utterance, wherein the context overlay cue defines a time period;
retrieving an episode frame tree set based on the time period;
retrieving an overlay rule template for the first persona of the RA/BE;
updating the current context of the RA/BE by overlaying the episode frame tree set based on the overlay rule template; and
performing at least one action based on the intents/entities of the first user utterance and the current context of the RA/BE.

2. The system of claim 1 , wherein, to retrieve the episode frame tree set, the processor is configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
retrieving a service-based frame of the episode frame tree set, wherein the service-based frame includes information retrieved from an external service.

3. The system of claim 1 , wherein the RA/BE comprises a second persona, and wherein, to retrieve the episode frame tree set, the processor is configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
retrieving a remote persona-based frame of the episode frame tree set, wherein the remote persona-based frame includes information retrieved from the second persona.

4. The system of claim 1 , wherein, to update the current context, the processor is configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
retrieving a plurality of episode frame tree sets based on the time period;
aggregating the plurality of episode frame tree sets based on multi-episode aggregation rules of the overlay rule template to generate an aggregate context; and
updating the current context of the RA/BE by overlaying the aggregate context based on the overlay rule template.

5. The system of claim 1 , wherein the processor is configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
receiving intents/entities of a second user utterance; and
performing another at least one action based on the intents/entities of the second user utterance and the current context of the RA/BE.

6. The system of claim 1 , wherein the memory is configured to store a persona context database that includes the episode frame tree set, wherein the episode frame tree set includes an episode start time, and an episode end time, and a plurality of frames that each include a plurality of parameters.

7. The system of claim 6 , wherein the episode start time and the episode end time are determined heuristically based on a change in topic or a delay in communication.

8. The system of claim 6 , wherein the processor is configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising:
identifying the episode frame tree set associated with the time period based on the episode start time and the episode end time of the episode frame tree set.

9. The system of claim 1 , wherein the actions comprise actions defined within a script associated with the first persona of the RA/BE.

10. The system of claim 1 , wherein the actions comprise action blocks defined within a script associated with the first persona of the RA/BE.

11. The system of claim 1 , wherein the actions comprise a cross-persona invocation requesting another persona of the RA/BE to perform other actions.

12. A method of operating a reasoning agent/behavior engine (RA/BE) of an agent automation system, comprising:
generating an episode frame tree set based on previously received user utterances and storing the episode frame tree set in a persona context database, wherein the episode frame tree set comprises an episode start time and an episode end time that are heuristically determined from the previously received user utterances;
receiving intents/entities of a user utterance;
recognizing a context overlay cue in the intents/entities of the user utterance, wherein the context overlay cue defines a time period;
retrieving the episode frame tree set from the persona context database based on the time period;
retrieving an overlay rule template from an overlay rule template database of the RA/BE;
updating a current context of the RA/BE by overlaying the episode frame tree set based on the overlay rule template; and
performing actions in response to the received user utterance based on the current context of the RA/BE.

13. The method of claim 12 , wherein retrieving the episode frame tree set comprises:
retrieving a plurality of episode frame tree sets, including the episode frame tree set, from the persona context database based on the time period.

14. The method of claim 13 , wherein updating the current context of the RA/BE comprises:
aggregating the plurality of overlaying the episode frame tree sets to generate an aggregate context based one or more multi-episode aggregation rules of the overlay rule template; and
updating the current context of the RA/BE by overlaying the aggregate context based on the overlay rule template.

15. A non-transitory, computer-readable medium storing instructions of an agent automation system executable by one or more processors of a computing system, wherein the instructions comprise:
instructions to receive intents/entities of a user utterance;
instructions to recognize a context overlay cue in the intents/entities of the user utterance, wherein the context overlay cue defines a time period;
instructions to retrieve an episode frame tree set from a persona context database based on the time period, wherein the episode frame tree set includes an episode start time, and an episode end time, and a plurality of frames that each include a plurality of parameters;
instructions to retrieve an overlay rule template from an overlay rule template database associated with the agent automation system;
instructions to update a current context by overlaying the episode frame tree set based on the overlay rule template; and
instructions to perform actions in response to the intents/entities of the user utterance based on the current context.

16. The medium of claim 15 , wherein the instructions to retrieve the episode frame tree set, comprise:
instructions to retrieve only a temporal scoped portion, an execution scoped portion, a component-type scoped portion, or an aspect-level scoped portion of the episode frame tree set from the persona context database.

17. The medium of claim 15 , wherein the instructions to retrieve the episode frame tree set comprises instructions to retrieve a plurality of episode frame tree sets, including the episode frame tree set, from the persona context database based on the time period, and wherein the instructions to update the current context comprise:
instructions to aggregate the plurality of overlaying the episode frame tree sets to generate an aggregate context based one or more multi-episode aggregation rules of the overlay rule template; and
instructions to update t

Assignees

• Servicenow Inc

Inventors

• Sapugay Edwin
• Madamala Anil Kumar
• Naboka Maxim
• Sunkara Srinivas Satyasai
• Santos Lewis Savio Landry
• Subbarao Murali B

Classifications

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F40/295

  Named entity recognition · CPC title

• G06F40/35 Primary

  Discourse or dialogue representation · CPC title