Method and system for domain-optimized semantic tagging and task execution using task classification encoding

• US 8,478,584 B1
• Filed: 11/06/2012
• Issued: 07/02/2013
• Est. Priority Date: 11/06/2012
• Status: Expired due to Fees

• Current Assignee: Ask Ziggy Inc.
• Original Assignee: Ask Ziggy Inc.
• Inventors: Grove, Kyle Wade
• Primary Examiner(s): Harper, Vincent P
• Application Number:
  US13/669,796
• Time in Patent Office: 238 Days
• Field of Search: 704/9, 704/3, 704/4, 704/275
• US Class Current: 704/9
• CPC Class Codes: G06F 40/126 Character encoding
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation

Abstract

A method and system for performing a task in a natural language processing system. The method includes processing text to obtain a numerical encoding, where the text includes a number of entities, where at least one of the entities is tagged with the numerical encoding, and where the text is derived from an utterance. The method further includes determining that the numerical encoding identifies a task, determining a variable required to complete the task using the numerical encoding, and generating a first dialogue sentence to obtain a variable value for the variable, where the variable is associated with a second numerical encoding, where the second numerical encoding includes the numerical encoding. The method further includes transmitting the first dialogue sentence to a user device, receiving the variable value for the variable in response to the first dialogue sentence, and performing the task using the variable value.

Claims

1. A method for performing a task in a natural language processing (NLP) system, comprising:

processing text to obtain a plurality of numerical encodings, wherein the text comprises a plurality of entities, wherein each of the plurality of numerical encodings is associated with one of the plurality of entities, and wherein the text is derived from an utterance;

populating a task list with each of the plurality of numerical encodings that have a length less than a pre-determined length;

populating a variable list with each of the plurality of numerical encodings that have a length greater than a pre-determined length;

selecting a numerical encoding from the task list, wherein the numerical encoding identifies a task in a domain-specific schema;

determining that the variable list does not include a variable value required to perform the task;

generating a sentence to obtain the variable value associated with a variable, wherein the variable is associated with a second numerical encoding, and wherein the second numerical encoding includes the numerical encoding;

transmitting the sentence to a user device;

in response to the transmitting, receiving the variable value for the variable in response to the dialogue sentence; and

performing the task using the variable value.

2. The method of claim 1, further comprising:

receiving user audio packets comprising the utterance from the user device; and

converting the user audio packets to text.

3. The method of claim 1, wherein each of the numerical encodings in the task list identifies one selected from a group consisting of a task element in the domain-specific schema and a sub-task element in the domain-specific schema.

4. The method of claim 1, wherein each of the numerical encodings in the variable list identifies one selected from a group consisting of one of a plurality of variable groups in the domain-specific schema and one of plurality of variables in the domain-specific schema, wherein the variable is one of the plurality of variables.

5. A method for performing tasks in a natural language processing (NLP) system, comprising:

processing text to obtain a plurality of numerical encodings, wherein the text comprises a plurality of entities, wherein each of the plurality of numerical encodings is associated with one of the plurality of entities, and wherein the text is derived from an utterance;

populating a task list with each of the plurality of numerical encodings that have a length less than a pre-determined length;

populating a variable list with each of the plurality of numerical encodings that have a length greater than a pre-determined length;

selecting a numerical encoding from the task list, wherein the numerical encoding identifies a sub-task in a domain-specific schema;

determining a task associated with the sub-task using the numerical encoding, wherein the task is associated with a second numerical encoding and wherein the numerical encoding includes the second numerical encoding;

determining a second sub-task associated with the task, wherein the second sub-task is associated with a third numerical encoding and wherein the third numerical encoding includes the second numerical encoding; and

performing the sub-task and the second sub-task.

6. The method of claim 5, wherein the sub-task and the second sub-task are performed in parallel.

7. The method of claim 5, wherein performing the sub-task comprises:

identifying a variable required to perform the sub-task;

identifying a numerical encoding associated with the variable; and

making a determination that the numerical encoding associated with the variable is in the variable list,

wherein performing the sub-task comprises using an entity of the plurality of entities, wherein the entity is associated with the numerical encoding associated with the variable.

8. The method of claim 5, wherein performing the sub-task comprises:

identifying a variable group required to perform the sub-task, wherein the variable group is associated with a plurality of variables;

making a determination that the variable list does not include any numerical encoding associated with any of the plurality of variables;

selecting a sentence that includes a second numerical encoding, wherein the second numerical encoding is associated with a variable group;

generating complete sentence using the sentence and a variable group name associated with the variable group;

transmitting the complete sentence to a user device; and

in response to the transmitting, receiving a variable value, wherein the variable value is associated with one of the plurality of variables,

wherein performing the sub-task comprises using the variable value.

9. The method of claim 8, wherein each of the plurality of variables is associated with a distinct numerical encoding, wherein each of the distinct numerical encodings includes the second numerical encoding.