Granted Patent B2

US 12,511,139 · App. 17/841,766 · Granted Dec 30, 2025

Inventors: Arijit Sengupta (San Francisco, CA); Jonathan Wray (San Francisco, CA); Grigory Nudelman (San Francisco, CA); Daniel Kane (San Francisco, CA); Geoffrey Grant (San Francisco, CA)

Assignee: Aible Inc.

G06F9/453G06F3/0482G06N5/04G06Q10/06312G06Q10/0633

Abstract

Claims (56)

1 . A method comprising:

receiving historical user behavior including historical data analysis characteristics;

generating, based on the historical data analysis characteristics, a file or set of files specifying a description of a bias field tag, wherein the bias field tag is indicative of a discriminatory bias associated with a data field in the generated file or set of files;

receiving a request to generate at least one of a predictive model, an optimization model, and/or a prescriptive model, and a task description for the predictive model, the optimization model, and/or the prescriptive model;

providing, on a user interface, a recommendation for generating the predictive model, the optimization model, and/or the prescriptive model based at least on the bias field tag;

receiving, via the user interface, user feedback on the provided recommendation; and

generating, using the file or set of files and the received user feedback, the predictive model, the optimization model, and/or the prescriptive model, the generating including training the predictive model, the optimization model, and/or the prescriptive model with a dataset.

2 . The method of claim 1 , wherein the historical user behavior is obtained from identified enterprise integrations.

3 . The method of claim 1 , wherein the file or set of files specify user interface configuration parameters, the user interface configuration parameters including an automated variable mapping that characterizes variable headers of a dataset and maps the variable headers to a specific explanatory variable in the file or set of files.

4 . The method of claim 1 , wherein the file or set of files specify user interface configuration parameters, the user interface configuration parameters including an actionability tag identifying a variable to change to influence an outcome; the method further comprising:

generating a recommendation based on the actionability tag.

5 . The method of claim 1 , wherein the file or set of files specify user interface configuration parameters, the user interface configuration parameters including a tag for a directional preference for an explanatory variable; the method further comprising:

generating a recommendation based on the tag.

6 . The method of claim 1 , further comprising determining data analysis characteristics, the determining the data analysis characteristics includes identifying variables in a dataset, identifying data sources to form the dataset, identifying a data granularity, and identifying rows and/or columns in the dataset to include and/or exclude.

7 . The method of claim 1 , wherein the historical data analysis characteristics includes metadata characterizing historical variables used; derived variables; historical data sources used; historical data granularity; historical data columns and/or rows to exclude and/or include; historical tasks performed by users; or a combination thereof.

8 . The method of claim 7 , wherein the generating includes using a model trained with the metadata to automatically generate a blueprint.

9 . The method of claim 7 , wherein the data analysis characteristics include a variable, a data source, a data granularity, a data column and/or row to exclude and/or include, and historical tasks performed by users.

10 . The method of claim 9 , the generating the predictive model, the optimization model, and/or the prescriptive model including training the model with a dataset according to the data analysis characteristics, the dataset including the variable obtained from the data source at the data granularity.

11 . The method of claim 7 , further comprising:

determining a performance of the predictive model, the optimization model, and/or the prescriptive model, the determining including determining a first performance value of the predictive model, the optimization model, and/or the prescriptive model;

rendering, within a graphical user interface, a plot including a first axis and a second axis, the first axis including a characterization of a first performance metric and the second axis including a characterization of a second performance metric; and

rendering, within the graphical user interface and the plot, a first graphical object at a first location characterizing the first performance value.

12 . The method of claim 11 , further comprising:

rendering a first line indicative of random model performance, a second line indicative of constant accuracy, and/or a third line indicative of constant cost,

wherein the first performance metric includes rate of false positive, count of false positive, cost of false positive, benefit missed by false positive, true positive, benefit of true positive, benefit of minimizing false positive, benefit of maximizing true positive, or a combination thereof;

wherein the second performance metric includes rate of false negative, count of false negative, cost of false negative, benefit missed by false negative, true negative, benefit of true negative, benefit of minimizing false negative, benefit of maximizing true negative, or a combination thereof.

13 . The method of claim 12 , further comprising:

rendering a first line indicative of target accuracy, a second line indicative of constant accuracy, and a third line indicative of constant cost; and receiving data characterizing a target accuracy; rendering a region indicative of the target accuracy, the region indicative of the target accuracy bounded by at least:

the first line indicative of the target accuracy and an origin of the plot;

a second line indicative of constant accuracy and the origin; or

a second line indicative of constant accuracy, the third line indicative of constant cost, and the origin;

wherein the first performance metric includes rate of false positive, count of false positive, cost of false positive, benefit missed by false positive, true positive, benefit of true positive, benefit of minimizing false positive, benefit of maximizing true positive, or a combination thereof;

wherein the second performance metric includes rate of false negative, count of false negative, cost of false negative, benefit missed by false negative, true negative, benefit of true negative, benefit of minimizing false negative, benefit of maximizing true negative, or a combination thereof.

14 . The method of claim 1 , wherein the generated file or set of files further comprises a description of at least one of an outcome variable, a use case specific text, a use case specific process flow or description.

15 . The method of claim 14 , wherein the file or set of files further specify user interface configuration parameters.

16 . The method of claim 15 , wherein the configuration parameters include a type of intended use case; a type of an outcome variable; an explanatory variable; a type of the explanatory variable; a data storage location for the explanatory variable; an automated variable mapping; an automated creation of variables; an automated variable classification; a use case specific input option; an actionability tag; an indication of positive and/or negative impact on the explanatory variable; a variable specific language tag; a variable specific language scale; or a combination thereof.

17 . A system comprising:

at least one data processor; and

memory storing instructions, which when executed by the at least one data processor, cause the at least one data processor to perform operations comprising:

receiving historical user behavior including historical data analysis characteristics;

generating, based on the historical data analysis characteristics, a file or set of files specifying a description of a bias field tag, wherein the bias field tag is indicative of a discriminatory bias associated with a data field in the generated file or set of files;

receiving a request to generate at least one of a predictive model, an optimization model, and/or a prescriptive model, and a task description for the predictive model, the optimization model, and/or the prescriptive model;

providing, on a user interface, a recommendation for generating the predictive model, the optimization model, and/or the prescriptive model based at least on the bias field tag;

receiving, via the user interface, user feedback on the provided recommendation; and

generating, using the file or set of files and the received user feedback, the predictive model, the optimization model, and/or the prescriptive model, the generating including training the predictive model, the optimization model, and/or the prescriptive model with a dataset.

18 . The system of claim 17 , wherein the historical user behavior is obtained from identified enterprise integrations.

19 . The system of claim 17 , wherein the generated file or set of files further comprises a description of at least one of an outcome variable, a use case specific text, a use case specific process flow or description.

20 . The system of claim 19 , wherein the file or set of files specify user interface configuration parameters.

21 . The system of claim 20 , wherein the configuration parameters include a type of intended use case; a type of an outcome variable; an explanatory variable; a type of the explanatory variable; a data storage location for the explanatory variable; an automated variable mapping; an automated creation of variables; an automated variable classification; a use case specific input option; an actionability tag; an indication of positive and/or negative impact on the explanatory variable; a variable specific language tag; a variable specific language scale; or a combination thereof.

22 . A non-transitory computer readable medium storing instructions which, when executed by at least one data processor, cause the at least one data processor to perform operations comprising:

receiving historical user behavior including historical data analysis characteristics;

generating, based on the historical data analysis characteristics, a file or set of files specifying a description of a bias field tag, wherein the bias field tag is indicative of a discriminatory bias associated with a data field in the generated file or set of files;

receiving a request to generate at least one of a predictive model, an optimization model, and/or a prescriptive model, and a task description for the predictive model, the optimization model, and/or the prescriptive model;

providing, on a user interface, a recommendation for generating the predictive model, the optimization model, and/or the prescriptive model based at least one the bias field tag;

receiving, via the user interface, user feedback on the provided recommendation; and

generating, using the file or set of files and the received user feedback, the predictive model, the optimization model, and/or the prescriptive model, the generating including training the predictive model, the optimization model, and/or the prescriptive model with a dataset.

Assignments (1)

Continuity (4)

References Cited (70)