Augmented natural language generation platform

Patent metadata

Field	Value
Publication number	US-12164864-B2
Application number	US-202318336417-A
Country	US
Kind code	B2
Filing date	Jun 16, 2023
Priority date	Sep 30, 2019
Publication date	Dec 10, 2024
Grant date	Dec 10, 2024

Abstract

A computing system receives a request to generate one or more narrative frameworks for a worksheet. The worksheet is hosted by a third party server. The computing system interfaces with the third party server to retrieve data associated with the worksheet. The computing system infers attributes of the worksheet from the retrieved data. The attributes correspond to a type of data represented in the worksheet. The computing system generates a narrative template for the one or more narrative frameworks. The computing system generates one or more data variables for the one or more narrative frameworks based on the generated narrative template. The computing system constructs the one or more narrative frameworks by selectively retrieving data from the worksheet for each data variable of the one or more data variables in the narrative framework.

First claim

The invention claimed is:

1. A method, comprising:
receiving, by a computing system, a first request to generate a narrative description based on information in a worksheet, the narrative description comprising a phrase that includes the information from the worksheet;
inferring, by the computing system and the worksheet, attributes corresponding to a type of data represented in the worksheet;
identifying, by the computing system, a subset of motifs from a plurality of motifs relevant to the inferred attributes of the worksheet;
generating, by the computing system, one or more data variables and a template comprising the one or more data variables based on the subset of motifs;
receiving, by the computing system, a second request to modify a value of a data variable of the one or more data variables in the narrative description;
updating, by the computing system, the data variable of the one or more data variables based on the second request;
extracting, by the computing system, data values from the worksheet, the data values corresponding to the one or more data variables; and
constructing, by the computing system, the narrative description by inputting the data values into the one or more data variables of the template to form the phrase of the narrative description.

2. The method of claim 1 , wherein inferring, by the computing system, the attributes of the worksheet comprises:
inferring data column types for the worksheet.

3. The method of claim 1 , further comprising:
causing, by the computing system, display of the narrative description in a graphical user interface.

4. The method of claim 3 , further comprising:
receiving, by the computing system via the graphical user interface, user input associated with a format of a first value in a first data variable; and
updating, by the computing system, the narrative description based on the format.

5. The method of claim 1 , wherein identifying, by the computing system, the subset of motifs from the plurality of motifs relevant to the inferred attributes of the worksheet comprises:
comparing the plurality of motifs against a collection of templates to identify the subset of motifs relevant to the inferred attributes of the worksheet.

6. The method of claim 1 , wherein inferring, by the computing system, the attributes of the worksheet comprises:
inferring data column types; or
inferring metadata associated with columns and tables in the data content.

7. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by one or more processors, causes a computing system to perform operations comprising:
receiving, by a computing system, a first request to generate a narrative description based on information in a worksheet, the narrative description comprising a phrase that includes the information from the worksheet;
inferring, by the computing system and from the worksheet, attributes corresponding to a type of data represented in the worksheet;
identifying, by the computing system, a subset of motifs from a plurality of motifs relevant to the inferred attributes of the worksheet;
generating, by the computing system, one or more data variables and a template comprising the one or more data variables based on the subset of motifs;
receiving, by the computing system, a second request to modify a value of a data variable of the one or more data variables in the narrative description;
updating, by the computing system, the data variable of the one or more data variables based on the second request;
extracting, by the computing system, data values from the worksheet, the data values corresponding to the one or more data variables; and
constructing, by the computing system, the narrative description by inputting the data values into the one or more data variables of the template to form the phrase of the narrative description.

8. The non-transitory computer readable medium of claim 7 , wherein inferring, by the computing system, the attributes of the worksheet from the data content comprises:
inferring data column types for the worksheet.

9. The non-transitory computer readable medium of claim 7 , wherein the sequences of instructions, when executed by the one or more processors, causes the computing system to perform operations comprising:
causing, by the computing system, display of the narrative description in a graphical user interface.

10. The non-transitory computer readable medium of claim 9 , wherein the sequences of instructions, when executed by the one or more processors, causes the computing system to perform operations comprising:
receiving, by the computing system via the graphical user interface, user input associated with a format of a first value in a first data variable; and
updating, by the computing system, the narrative description based on the format.

11. The non-transitory computer readable medium of claim 7 , wherein identifying, by the computing system, the subset of motifs from the plurality of motifs relevant to the inferred attributes of the worksheet comprises:
comparing the plurality of motifs against a collection of templates to identify the subset of motifs relevant to the inferred attributes of the worksheet.

12. The non-transitory computer readable medium of claim 7 , wherein inferring, by the computing system, the attributes of the worksheet comprises:
inferring data column types; or
inferring metadata associated with columns and tables.

13. A system comprising:
a processor; and
a memory having programming instructions stored thereon, which, when executed by the processor, cause the system to perform operations comprising:
receiving a first request to generate a narrative description based on information in a worksheet, the narrative description comprising a phrase that includes the information from the worksheet;
inferring attributes of the worksheet from the data content, the attributes corresponding to a type of data represented in the worksheet;
identifying a subset of motifs from a plurality of motifs relevant to the inferred attributes of the worksheet;
generating one or more data variables and a template comprising the one or more data variables based on the subset of motifs;
receiving a second request to modify a value of a data variable of the one or more data variables in the narrative description;
updating the data variable of the one or more data variables based on the second request;

extracting data values from the worksheet, the data values corresponding to the one or more data variables; and
constructing the narrative description by inputting the data values into the one or more data variables of the template to form the phrase of the narrative description.

14. The system of claim 13 , wherein inferring the attributes of the worksheet comprises:
inferring data column types for the worksheet.

15. The system of claim 13 , wherein the programming instructions, when executed by the processor, cause the system to perform operations comprising:
causing display of the narrative description in a graphical user interface;
receiving, via the graphical user interface, user input associated with a format of a first value in a first data variable; and
updating the narrative description based on the format.

16. The system of claim 13 , wherein identifying the subset of motifs from the plurality of motifs relevant to the inferred attributes of the worksheet comprises:
comparing the plurality of motifs against a collection of templates to identify the subset of motifs relevant to the inferred attri

Assignees

• Stats Llc

Inventors

• Long Adam
• Haynes Nick
• Salo Dan
• Dahm Heather
• Mccright Jacob
• Dillon Michael
• Deters Hannah
• Khadka Anup

Classifications

• G06F40/205

  Parsing · CPC title

• G06F40/56

  Natural language generation · CPC title

• G06F40/186 Primary

  Templates · CPC title