Using Unstructured Input to Update Heterogeneous Data Stores

Patent metadata

Field	Value
Publication number	US-2020089757-A1
Application number	US-201816134956-A
Country	US
Kind code	A1
Filing date	Sep 18, 2018
Priority date	Sep 18, 2018
Publication date	Mar 19, 2020
Grant date	—

Abstract

Approaches to using unstructured input to update heterogeneous data stores include receiving unstructured text input, receiving a template for interpreting the unstructured text input, identifying, using an entity classifier, entities in the unstructured text input, identifying one or more potential parent entities from the identified entities based on the template, receiving a selection of a parent entity from the one or more potential parent entities, identifying one or more potential child entities from the identified entities based on the template and the selected parent entity, receiving a selection of a child entity from the one or more potential child entities, identifying an action item in the unstructured text input based on the identified entities and the template, determining, using an intent classifier, an intent of the action item, and updating a data store based on the determined intent, the identified entities, and the selected child entity.

First claim

What is claimed is:

1 . A method for natural language processing, the method comprising:
receiving, by one or more processors of a server, unstructured text input; receiving, by the one or more processors, a template for interpreting the unstructured text input; identifying, using an entity classifier, entities in the unstructured text input; identifying, by the one or more processors, one or more potential parent entities from the identified entities based on the template; receiving, by the one or more processors, a selection of a parent entity from the one or more potential parent entities; identifying, by the one or more processors, one or more potential child entities from the identified entities based on the template and the selected parent entity; receiving, by the one or more processors, a selection of a child entity from the one or more potential child entities; identifying, by the one or more processors, an action item in the unstructured text input based on the identified entities and the template; determining, using an intent classifier, an intent of the action item; and updating a data store based on the determined intent, the identified entities, and the selected child entity.

2 . The method of claim 1 , wherein updating the date store is further based on the selected parent entity.

3 . The method of claim 1 , wherein identifying the entities comprises identifying a type of each of the entities, a value of each of the entities, and a confidence level in the identifying.

4 . The method of claim 3 , wherein a type of each of the entities is selected from a group consisting of an organization, a person, a date, a time, a percentage, a monetary value, and a pick list type.

5 . The method of claim 1 , wherein updating the data store based on the determined intent, the identified entities, and the selected child entity comprises:
matching the determined intent and types of one or more entities identified in the action item to an action item signature in the template; and generating a database query based on the signature and values of the one or more entities identified in the action item.

6 . The method of claim 5 , wherein a field from a record in the data store to include in the database query is identified in the template by a type of a corresponding entity identified in the action item.

7 . The method of claim 1 , wherein:
identifying the one or more potential parent entities comprises searching in records of one or more parent tables in the data store for values matching one or more of the identified entities having a type matching a type of the one or more parent tables; and the one or more parent tables are identified in the template.

8 . The method of claim 1 , wherein:
identifying the one or more potential child entities comprises searching in records of one or more child tables in the data store for values matching one or more of the identified entities having a type matching a type of the one or more child tables; the one or more child tables are identified in the template; and the records are associated with a record corresponding to the selected parent entity.

9 . The method of claim 8 , wherein identifying the one or more potential child entities further comprises one or more of:
filtering the records based on a filter in the template; or ordering the one or more potential child entities based on an ordering specified by the template.

10 . The method of claim 1 , further comprising publishing changes to the data store based on the updating.

11 . The method of claim 1 , wherein:
the unstructured text input is received as audio input; and the method further comprises performing speech recognition on the audio input.

12 . A non-transitory machine-readable medium comprising executable code which when executed by one or more processors associated with a computing device are adapted to cause the one or more processors to perform a method comprising:
receiving unstructured text input; receiving a template for interpreting the unstructured text input; identifying, using an entity classifier, entities in the unstructured text input; identifying one or more potential parent entities from the identified entities based on the template; receiving a selection of a parent entity from the one or more potential parent entities; identifying one or more potential child entities from the identified entities based on the template and the selected parent entity; receiving a selection of a child entity from the one or more potential child entities; identifying an action item in the unstructured text input based on the identified entities and the template; determining, using an intent classifier, an intent of the action item; and updating a data store based on the determined intent, the identified entities, and the selected child entity.

13 . The non-transitory machine-readable medium of claim 12 , wherein a type of each of the entities is selected from a group consisting of an organization, a person, a date, a time, a percentage, a monetary value, and a pick list type.

14 . The non-transitory machine-readable medium of claim 12 , wherein updating the data store based on the determined intent, the identified entities, and the selected child entity comprises:
matching the determined intent and types of one or more entities identified in the action item to an action item signature in the template; and generating a database query based on the signature and values of the one or more entities identified in the action item.

15 . The non-transitory machine-readable medium of claim 14 , wherein a field from a record in the data store to include in the database query is identified in the template by a type of a corresponding entity identified in the action item.

16 . The non-transitory machine-readable medium of claim 12 , wherein:
identifying the one or more potential parent entities comprises searching in records of one or more parent tables in the data store for values matching one or more of the identified entities having a type matching a type of the one or more parent tables; and the one or more parent tables are identified in the template.

17 . The non-transitory machine-readable medium of claim 12 , wherein:
identifying the one or more potential child entities comprises searching in records of one or more child tables in the data store for values matching one or more of the identified entities having a type matching a type of the one or more child tables; the one or more child tables are identified in the template; and the records are associated with a record corresponding to the selected parent entity.

18 . A computing device comprising:
a memory; and one or more processors coupled to the memory; wherein the one or more processors are configured to:
receive unstructured text input;
receive a template for interpreting the unstructured text input;
identify, using an entity classifier, entities in the unstructured text input;
identify, one or more potential parent entities from the identified entities based on the template;
receive a selection of a parent entity from the one or more potential parent entities;
identify one or more potential child entities from the identified entities based on the template and the selected parent entity;
receive a selection of a child entity from the one or more potential child entities;
identify an action item in the unstructured text input based on the identified entities and the template;
determine, using an intent

Assignees

• Salesforce Com Inc

Inventors

• Machado Michael
• Ball John
• Cook Jr Thomas Archie
• Harinath Shashank
• Lalani Roojuta
• Laraki Zineb
• Liu Qingqing
• Rosenbaum Mike
• Skucha Karl Ryszard
• Soumet Jean-Marc
• Vijayakumar Manju

Classifications

• G06F16/3329

  Natural language query formulation · CPC title

• G10L15/1815

  Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F40/295 Primary

  Named entity recognition · CPC title

• G06F40/205

  Parsing · CPC title