Providing an object-based response to a natural language query

Patent metadata

Field	Value
Publication number	US-12353477-B2
Application number	US-202418673111-A
Country	US
Kind code	B2
Filing date	May 23, 2024
Priority date	Dec 10, 2018
Publication date	Jul 8, 2025
Grant date	Jul 8, 2025

Abstract

A data analysis system presents a user interface to allow a user to provide a natural language query pertaining to a dataset, wherein the dataset is associated with a data object model comprising a plurality of objects and receives, via the user interface, user input specifying the natural language query. The data analysis system further modifies, in the user interface, the user input to visually indicate one or more portions of the natural language query that each represent one of the plurality of objects and presents, in the user interface, a response to the natural language query, the response being based on data from the dataset, the data corresponding to the one of the plurality of objects.

First claim

What is claimed is:

1. A method comprising:
receiving a data string comprising a natural language query pertaining to a dataset, the dataset being associated with a data object model comprising a plurality of objects;
parsing the data string to identify a plurality of individual words within the data string;
applying the plurality of individual words as an input to a trained machine learning model to obtain an output, the output including an indication of one or more objects of the plurality of objects;
generating a response to the natural language query based at least in part on the one or more objects; and
determining one or more artifacts using the trained machine learning model based on the dataset, wherein each artifact of the one or more artifacts is associated with at least one of the one or more objects;
wherein the method is performed using one or more processors.

2. The method of claim 1 , further comprising:
determining whether one or more individual words of the plurality of individual words correspond to a definition of the one or more objects included within an ontology associated with the data object model.

3. The method of claim 1 , wherein the output includes an indication of the one or more artifacts.

4. The method of claim 3 , wherein the output further includes a dynamic relevance score for each artifact of the one or more artifacts.

5. The method of claim 4 , wherein the dataset includes a plurality of datasets, wherein the method further comprises:
selecting an artifact from the one or more artifacts having a highest dynamic relevance score;
identifying a corresponding dataset in the plurality of datasets from an artifact index; and
executing the selected artifact against the corresponding dataset to generate the response to the natural language query.

6. The method of claim 1 , further comprising:
identifying previously stored natural language queries that are similar to the natural language query in the data string; and
providing at least one of the previously stored natural language queries as an alternative query.

7. The method of claim 1 , wherein the trained machine learning model is trained using a plurality of historical natural language queries and historical responses.

8. The method of claim 1 , further comprising:
presenting a user interface to allow a user to provide the natural language query;
receiving a user input indicating a first command corresponding to the response, the first command causing the response to the natural language query to be recreated until a second command is received.

9. The method of claim 8 , further comprising:
modifying, in the user interface, a visual indication of one portion of the natural language query to a selectable interface element including a plurality of selectable options associated with the one portion of the natural language query, wherein the selectable interface element is part of a visual indication of the natural language query.

10. A system comprising:
one or more memories; and
one or more processors coupled to the one or more memories and configured to perform operations comprising:
receiving a data string comprising a natural language query pertaining to a dataset, the dataset being associated with a data object model comprising a plurality of objects;
parsing the data string to identify a plurality of individual words within the data string;
applying the plurality of individual words as an input to a trained machine learning model to obtain an output, the output including an indication of one or more objects of the plurality of objects;
generating a response to the natural language query based at least in part on the one or more objects; and
determining one or more artifacts using the trained machine learning model based on the dataset, wherein each artifact of the one or more artifacts is associated with at least one of the one or more objects.

11. The system of claim 10 , wherein the operations further comprise:
determining whether one or more individual words of the plurality of individual words correspond to a definition of the one or more objects included within an ontology associated with the data object model.

12. The system of claim 10 , wherein the output includes an indication of the one or more artifacts.

13. The system of claim 12 , wherein the output further includes a dynamic relevance score for each artifact of the one or more artifacts;
wherein the dataset includes a plurality of datasets;
wherein the operations further comprise:
selecting an artifact from the one or more artifacts having a highest dynamic relevance score;
identifying a corresponding dataset in the plurality of datasets from an artifact index; and
executing the selected artifact against the corresponding dataset to generate the response to the natural language query.

14. The system of claim 10 , wherein the operations further comprise:
identifying previously stored natural language queries that are similar to the natural language query in the data string; and
providing at least one of the previously stored natural language queries as an alternative query.

15. The system of claim 10 , wherein the trained machine learning model is trained using a plurality of historical natural language queries and historical responses.

16. The system of claim 10 , wherein the operations further comprise:
presenting a user interface to allow a user to provide the natural language query;
receiving a user input indicating a first command corresponding to the response, the first command causing the response to the natural language query to be recreated until a second command is received.

17. The system of claim 16 , wherein the operations further comprise:
modifying, in the user interface, a visual indication of one portion of the natural language query to a selectable interface element including a plurality of selectable options associated with the one portion of the natural language query, wherein the selectable interface element is part of a visual indication of the natural language query.

18. A non-transitory computer-readable storage medium storing instructions that, when executed by a processing device, cause the processing device to perform operations comprising:
receiving a data string comprising a natural language query pertaining to a dataset, the dataset being associated with a data object model comprising a plurality of objects;
parsing the data string to identify a plurality of individual words within the data string;
applying the plurality of individual words as an input to a trained machine learning model to obtain an output, the output including an indication of one or more objects of the plurality of objects;
generating a response to the natural language query based at least in part on the one or more objects; and
determining one or more artifacts using the trained machine learning model based on the dataset, wherein each artifact of the one or more artifacts is associated with at least one of the one or more objects.

Assignees

• Palantir Technologies Inc

Inventors

• Lisuk David
• Porter Eric
• Shashi Aditya
• Soloducho Ilai
• Wiseheart John
• Xu Guodong
• Foks Maciej

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06F40/205 Primary

  Parsing · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F16/90332

  Natural language query formulation or dialogue systems · CPC title