Granted Patent B1

US 12,547,633 · App. 18/746,830 · Granted Feb 10, 2026

Inventors: Jennifer Dawn Ayala (San Antonio, TX); Zachery C. Lake (Fort Wayne, IN); Sydney Ann Conrad-Cook (Van Meter, IA); Courtney St. Martin (Forney, TX)

Assignee: United Services Automobile Association (USAA)

G06F16/24575G06F16/2428G06F16/285

Abstract

Claims (49)

1 . A method for selecting data analytics task components using one or more prediction models, the method comprising:

accessing, by a computing system, one or more previously stored prediction models configured to process previous data analytics task parameters, the one or more previously stored prediction models comprising one or more machine learning models;

receiving, by the computing system, task parameters for a particular data analytics task;

predicting, by the computing system and by providing the task parameters for the particular data analytics task to the one or more previously stored prediction models, a set of the previous data analytics tasks that match the particular data analytics task;

selecting, by the computing system and using the set of matching previous data analytics tasks, a subset of common data tables that are associated with each of the set of the previous data analytics tasks that match the particular data analytics task; and

providing, by the computing system, the subset of common data tables in relation to the particular data analytics task.

2 . The method of claim 1 , wherein the one or more previously stored prediction models comprise one or more unsupervised machine learning models configured to cluster analytics tasks by task parameters.

3 . The method of claim 2 , wherein the one or more previously stored prediction models are configured to determine similarity between the task parameters for the particular data analytics task and the task parameters for the previous data analytics tasks such that the set of matching previous data analytics tasks comprise a similarity that meets or exceeds a similarity criteria.

4 . The method of claim 1 , wherein the one or more previously stored prediction models were trained on task parameters, for at least a portion of the previous data analytics tasks, that comprise a search query used to search one or more data tables.

5 . The method of claim 4 , wherein the parameters for the portion of the previous data analytics tasks comprise one or more attributes of the search query, the attributes including data tables, data columns, data table joins, data query functions, or any combination thereof.

6 . The method of claim 1 , further comprising:

receiving a selection for at least one of the subset of common data tables;

grouping additional parameters for the selected data table, wherein the grouped additional parameters comprise parameter values from a subset of the matching previous data analytics tasks, the subset comprising matching previous data analytics tasks that include the selected data table as a parameter; and

providing at least one of the grouped additional parameters.

7 . The method of claim 6 , wherein the at least one of the grouped additional parameters comprises a set of grouped task descriptions, wherein the set of grouped task descriptions comprise parameters for the subset of the matching previous data analytics tasks.

8 . The method of claim 6 , wherein the at least one of the grouped additional parameters comprises a set of grouped table joins, wherein the set of grouped table joins comprise parameters for the subset of the matching previous data analytics tasks.

9 . The method of claim 1 , further comprising:

retrieving a set of data tables that comprise parameters for the matching previous data analytics tasks;

calculating a commonality metric for each retrieved data table within the set, the commonality metric for a given table comprising the number of the matching previous data analytics tasks that comprise the given table as a parameter; and

comparing the commonality metric for each retrieved data table from the set to a criteria, wherein the subset of common data tables comprise the data tables from the set with a commonality metric that meets the criteria.

10 . A computer-readable storage medium storing instructions that, when executed by a computing system, cause the computing system to perform a process for selecting data analytics task components using one or more prediction models, the process comprising:

accessing, by a computing system, one or more previously stored prediction models configured to process previous data analytics task parameters, the one or more previously stored prediction models comprising one or more machine learning models;

receiving, by the computing system, task parameters for a particular data analytics task;

predicting, by the computing system and by providing the task parameters for the particular data analytics task to the one or more previously stored prediction models, a set of the previous data analytics tasks that match the particular data analytics task;

selecting, by the computing system and using the set of matching previous data analytics tasks, a subset of common data tables that are associated with each of the set of the previous data analytics tasks that match the particular data analytics task; and

providing, by the computing system, the subset of common data tables in relation to the particular data analytics task.

11 . The computer-readable storage medium of claim 10 , wherein the process further comprises:

retrieving a set of data tables that comprise parameters for the matching previous data analytics tasks;

calculating a commonality metric for each retrieved data table within the set, the commonality metric for a given table comprising the number of the matching previous data analytics tasks that comprise the given table as a parameter; and

comparing the commonality metric for each retrieved data table from the set to a criteria, wherein the subset of common data tables comprise the data tables from the set with a commonality metric that meets the criteria.

12 . The computer-readable storage medium of claim 10 , wherein the one or more previously stored prediction models comprise one or more unsupervised machine learning models configured to cluster analytics tasks by task parameters.

13 . The computer-readable storage medium of claim 12 , wherein the one or more previously stored prediction models are configured to determine similarity between the task parameters for the particular data analytics task and the task parameters for the previous data analytics tasks such that the set of matching previous data analytics tasks comprise a similarity that meets or exceeds a similarity criteria.

14 . The computer-readable storage medium of claim 10 , wherein the one or more previously stored prediction models were trained on task parameters, for at least a portion of the previous data analytics tasks, that comprise a search query used to search one or more data tables.

15 . The computer-readable storage medium of claim 14 , wherein the parameters for the portion of the previous data analytics tasks comprise one or more attributes of the search query, the attributes including data tables, data columns, data table joins, data query functions, or any combination thereof.

16 . The computer-readable storage medium of claim 10 , wherein the process further comprises:

receiving a selection for at least one of the subset of common data tables;

grouping additional parameters for the selected data table, wherein the grouped additional parameters comprise parameter values from a subset of the matching previous data analytics tasks, the subset comprising matching previous data analytics tasks that include the selected data table as a parameter; and

providing at least one of the grouped additional parameters.

17 . The computer-readable storage medium of claim 16 , wherein the at least one of the grouped additional parameters comprises a set of grouped task descriptions, wherein the set of grouped task descriptions comprise parameters for the subset of the matching previous data analytics tasks.

18 . The computer-readable storage medium of claim 16 , wherein the at least one of the grouped additional parameters comprises a set of grouped table joins, wherein the set of grouped table joins comprise parameters for the subset of the matching previous data analytics tasks.

19 . A computing system for selecting data analytics task components using one or more prediction models, the computing system comprising:

one or more processors; and

one or more memories storing instructions that, when executed by the one or more processors, cause the computing system to perform a process comprising:

accessing, by a computing system, one or more previously stored prediction models configured to process previous data analytics task parameters, the one or more previously stored prediction models comprising one or more machine learning models;

receiving, by the computing system, task parameters for a particular data analytics task;

predicting, by the computing system and by providing the task parameters for the particular data analytics task to the one or more previously stored prediction models, a set of the previous data analytics tasks that match the particular data analytics task;

selecting, by the computing system and using the set of matching previous data analytics tasks, a subset of common data tables that are associated with each of the set of the previous data analytics tasks that match the particular data analytics task; and

providing, by the computing system, the subset of common data tables in relation to the particular data analytics task.

20 . The computing system of claim 19 , wherein the one or more previously stored prediction models were trained on task parameters, for at least a portion of the previous data analytics tasks, that comprise a search query used to search one or more data tables.

Assignments (2)

Continuity (1)

References Cited (73)