Granted Patent B1

US 12,511,282 · App. 18/653,504 · Granted Dec 30, 2025

Inventors: Ananya Ojha (Dunn Loring, VA); William Hurwood (Washington, DC)

Assignee: MicroStrategy Incorporated

G06F16/24522G06F16/2423G06F16/248

Abstract

Claims (61)

1 . A method performed by one or more computers, the method comprising:

providing, by the one or more computers, an interface configured to receive input comprising data processing criteria expressed in natural language text;

receiving, by the one or more computers, natural language input through the interface, wherein the natural language input specifies one or more criteria for retrieving or generating data from a data set;

determining that the data set is relevant to the natural language input; and

based on determining that the data set is relevant to the natural language input, selecting a data model for the data set;

obtaining, by the one or more computers, a structured query language (SQL) statement generated by an artificial intelligence or machine learning (AI/ML) model based on the natural language input, wherein obtaining the SQL statement comprises causing the AI/ML model to process the natural language input using the selected data model as context, and wherein the SQL statement expresses the one or more criteria from the natural language input using references to components of the data set or logical objects of the data set; and

providing, by the one or more computers, the SQL statement through the interface.

2 . The method of claim 1 , wherein the AI/ML model is a large language model.

3 . The method of claim 2 , wherein the large language model has been trained based on examples of computer-generated SQL statements and corresponding user-edited SQL statements made by editing the computer-generated SQL statements.

4 . The method of claim 1 , wherein obtaining the SQL statement comprises:

sending a request for the AI/ML model to generate the SQL statement, wherein the request includes:

the natural language input or a natural language instruction the AI/ML model that is generated based on the natural language input; and

data indicating contents or structure of the data set; and

receiving the SQL statement that the AI/ML model generated, in which elements of the natural language input are mapped to or translated to components of the data set.

5 . The method of claim 4 , wherein the data indicating the contents or structure of the data set comprises an object model or database schema for the data set.

6 . The method of claim 1 , wherein obtaining the SQL statement comprises generating an instruction to a large language model based on the natural language text.

7 . The method of claim 1 , wherein determining that the data set is relevant to the natural language input comprises:

comparing terms of the natural language input to data models for data sets or metadata for data sets; and

determining that the data set is relevant to the natural language input based on a match or similarity of the terms of the natural language input and the data model for the data sets or the metadata for the data set.

8 . The method of claim 1 , wherein determining that the data set is relevant to the natural language input comprises:

receiving data indicating an identifier for a data set used to provide content that is displayed in or selected on user interface content presented with a user interface in which the natural language input is entered; and

determining that the data set is relevant to the natural language input based on the received identifier.

9 . The method of claim 1 , comprising receiving the natural language input through an application programming interface (API).

10 . The method of claim 1 , comprising:

receiving the natural language input over a network from a user device; and

providing the SQL statement to the user device over a network for display in a user interface of a client device.

11 . The method of claim 1 , comprising:

providing the SQL statement to a database system;

obtaining results that the database system generates based on executing the SQL statement; and

providing at least a portion of the results or natural language text describing the results in response to the natural language input.

12 . The method of claim 1 , wherein the criteria specified by the natural language input comprise one or more of a data set from which to retrieve data, a type of data to retrieve, a selection criterion, a filter criterion, or a sorting criterion.

13 . The method of claim 1 , comprising:

saving the SQL statement at a server system;

creating an object representing the SQL statement or a set of results obtained from executing the SQL statement to obtain data from the data set;

causing a representation of the object to be provided in a user interface, wherein the representation of the object is interactive and configured to provide, upon interaction by a user, access to the SQL statement or a set of results generated by executing the SQL statement.

14 . The method of claim 13 , comprising forming an association between the object with the data set or a user, a document, or a project that is active when the natural language input is provided;

wherein the representation of the object is selectively caused to be presented based on the context of an application, wherein causing the representation of the object to be provided in the user interface is performed in response to determining that the user interface is associated with the data set or the user, the document, or the project.

15 . The method of claim 1 , wherein the data model indicates logical objects of the data set, including metrics or attributes that are stored in or can be derived from columns of data in the data set.

16 . A system comprising:

one or more computers; and

one or more non-transitory computer-readable media storing instructions that are operable, when executed by the one or more computers, to cause the system to perform operations comprising:

providing, by the one or more computers, an interface configured to receive input comprising data processing criteria expressed in natural language text;

receiving, by the one or more computers, natural language input through the interface, wherein the natural language input specifies one or more criteria for retrieving or generating data from a dataset;

determining that the data set is relevant to the natural language input; and

based on determining that the data set is relevant to the natural language input, selecting a data model for the data set;

obtaining, by the one or more computers, a structured query language (SQL) statement generated by an artificial intelligence or machine learning (AI/ML) model based on the natural language input, wherein obtaining the SQL statement comprises causing the AI/ML model to process the natural language input using the selected data model as context, and wherein the SQL statement expresses the one or more criteria from the natural language input using references to components of the data set or logical objects of the data set; and

providing, by the one or more computers, the SQL statement through the interface.

17 . The system of claim 16 , wherein the AI/ML model is a large language model.

18 . The system of claim 17 , wherein the large language model has been trained based on examples of computer-generated SQL statements and corresponding user-edited SQL statements made by editing the computer-generated SQL statements.

19 . The system of claim 16 , wherein obtaining the SQL statement comprises:

sending a request for the AI/ML model to generate the SQL statement, wherein the request includes:

the natural language input or a natural language instruction the AI/ML model that is generated based on the natural language input; and

data indicating contents or structure of the data set; and

receiving the SQL statement that the AI/ML model generated, in which elements of the natural language input are mapped to or translated to components of the data set.

20 . One or more non-transitory computer-readable media storing instructions that are operable, when executed by one or more computers, to cause the one or more computers to perform operations comprising:

providing, by the one or more computers, an interface configured to receive input comprising data processing criteria expressed in natural language text;

receiving, by the one or more computers, natural language input through the interface, wherein the natural language input specifies one or more criteria for retrieving or generating data from a data set;

determining that the data set is relevant to the natural language input; and

based on determining that the data set is relevant to the natural language input, selecting a data model for the data set;

obtaining, by the one or more computers, a structured query language (SQL) statement generated by an artificial intelligence or machine learning (AI/ML) model based on the natural language input, wherein obtaining the SQL statement comprises causing the AI/ML model to process the natural language input using the selected data model as context, and wherein the SQL statement expresses the one or more criteria from the natural language input using references to components of the data set or logical objects of the data set; and

providing, by the one or more computers, the SQL statement through the interface.

Assignments (2)

Continuity (2)

References Cited (143)

Cited By (2)

US 12,670,160 US 12,694,258