Interactive graphical user interface with override guidance

• US 10,255,085 B1
• Filed: 11/07/2018
• Issued: 04/09/2019
• Est. Priority Date: 03/13/2018
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Valsaraj, Varunraj, Aral, Bahadir, Yi, Jinxin, Baldridge, Jr., Roger Lee, Gallagher, Rebecca
• Primary Examiner(s): Shih, Haoshian
• Application Number:
  US16/182,930
• Time in Patent Office: 153 Days
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: G06F 9/453 Help systems
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/20 Ensemble learning
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound

Abstract

One exemplary system can receive a selection of a dataset via a graphical user interface (GUI). The dataset can represent a time-series projection. The system can feed the dataset into a first machine-learning model to obtain an output indicating whether the time-series projection has a data value that should be overridden with an override value. If the first machine-learning model indicates that the time-series projection has the data value that should be overridden, the system can feed the data value as input to a second machine-learning model to obtain an output indicating whether the override value should be greater than or less than the data value. The system can then render a visual directionality cue within the GUI based on the output from the second machine-learning model. The visual directionality cue can provide guidance for overriding the data value.

Claims

1. A system for generating a graphical user interface (GUI) with override guidance, the system comprising:

a processing device; and

a memory device including instructions that are executable by the processing device for causing the processing device to;

generate the GUI, wherein the GUI includes a first input component for receiving a first input indicating a dataset for display within the GUI;

receive the first input indicating the dataset via the first input component, wherein the dataset represents a time-series projection spanning a future time period, and wherein the time-series projection was previously generated by analyzing historical data spanning another time period that is prior to the future time period;

in response to receiving the first input, generate a graph of the time-series projection within the GUI;

automatically feed the dataset into a first machine-learning model to obtain an output from the first machine-learning model indicating whether the time-series projection has a data value that should be overridden with an override value;

in response to the output from the first machine-learning model indicating that the time-series projection has the data value that should be overridden;

automatically feed the data value as input to a second machine-learning model to obtain an output from the second machine-learning model indicating whether the override value should be greater than or less than the data value; and

automatically render a visual directionality cue within the GUI based on the output from the second machine-learning model, wherein the visual directionality cue is configured to provide guidance about whether the override value should be greater than or less than the data value;

receive a second input including the override value via a second input component of the GUI;

in response to receiving the second input, generate another version of the time-series projection in which the data value is overridden with the override value; and

render the other version of the time-series projection within the GUI.

2. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

generate teaching data using a plurality of time-series projections spanning one or more future time-periods, wherein the teaching data includes (i) predicted values for data points in the plurality of time-series projections, (ii) override values for the data points provided by a particular user, (iii) actual values for the data points gathered during the one or more future time-periods, and (iv) in indication of whether the override values are closer to the actual values than the predicted values.

3. The system of claim 2, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

train a plurality of machine-learning models using the teaching data to enable the plurality of machine-learning models to determine whether a particular data-value provided to the machine-learning model should be overridden;

determine accuracies of the plurality of machine-learning models using validation data;

compare the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

select the most accurate machine-learning model as the first machine-learning model.

4. The system of claim 2, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

train a plurality of machine-learning models using the teaching data and outputs from the first machine-learning model;

determine accuracies of the plurality of machine-learning models using validation data;

compare the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

select the most accurate machine-learning model as the second machine-learning model.

5. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

visually position a graphical indicator proximate to the data value in the graph in response to the first machine-learning model indicating that the data value should be overridden with the override value, wherein the graphical indicator indicates that the data value should be reviewed and is different from the visual directionality cue.

6. The system of claim 1, wherein the visual directionality cue is an arrow that is oriented (i) in a first direction to indicate that the override value should be greater than the data value or (ii) in a second direction that is opposite to the first direction to indicate that the override value should be less than the data value.

7. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

automatically determine an upper limit and a lower limit for the override value by analyzing the historical data used to generate the time-series projection; and

automatically render visual markers within the GUI indicating the upper limit and the lower limit for the override value.

8. The system of claim 7, wherein:

the GUI includes a data table having a column associated with the data value; and

the column includes (i) the data value in a row, (ii) the second input component for inputting the override value in another row, and (iii) the visual directionality cue in yet another row.

9. The system of claim 8, wherein the column further includes the upper limit and the lower limit in respective rows.

10. The system of claim 7, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

analyze the override value and visually display an error indicator within the GUI if the override value is out of a range defined by the upper limit and the lower limit.

11. A non-transitory computer-readable medium comprising program code that is executable by a processing device for causing the processing device to:

generate a graphical user interface (GUI) that includes a first input component for receiving a first input indicating a dataset for display within the GUI;

receive the first input indicating the dataset via the first input component, wherein the dataset represents a time-series projection spanning a future time period, and wherein the time-series projection was previously generated by analyzing historical data spanning another time period that is prior to the future time period;

in response to receiving the first input, generate a graph of the time-series projection within the GUI;

automatically feed the dataset into a first machine-learning model to obtain an output from the first machine-learning model indicating whether the time-series projection has a data value that should be overridden with an override value;

in response to the output from the first machine-learning model indicating that the time-series projection has the data value that should be overridden;

automatically feed the data value as input to a second machine-learning model to obtain an output from the second machine-learning model indicating whether the override value should be greater than or less than the data value; and

automatically render a visual directionality cue within the GUI based on the output from the second machine-learning model, wherein the visual directionality cue is configured to provide guidance about whether the override value should be greater than or less than the data value;

receive a second input including the override value via a second input component of the GUI;

in response to receiving the second input, generate another version of the time-series projection in which the data value is overridden with the override value; and

render the other version of the time-series projection within the GUI.

12. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to:

generate teaching data using a plurality of time-series projections spanning one or more future time-periods, wherein the teaching data includes (i) predicted values for data points in the plurality of time-series projections, (ii) override values for the data points provided by a particular user, (iii) actual values for the data points gathered during the one or more future time-periods, and (iv) in indication of whether the override values are closer to the actual values than the predicted values.

13. The non-transitory computer-readable medium of claim 12, further comprising program code that is executable by the processing device for causing the processing device to:

train a plurality of machine-learning models using the teaching data to enable the plurality of machine-learning models to determine whether a particular data-value provided to the machine-learning model should be overridden;

determine accuracies of the plurality of machine-learning models using validation data;

compare the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

select the most accurate machine-learning model as the first machine-learning model.

14. The non-transitory computer-readable medium of claim 12, further comprising program code that is executable by the processing device for causing the processing device to:

train a plurality of machine-learning models using the teaching data and outputs from the first machine-learning model;

determine accuracies of the plurality of machine-learning models using validation data;

compare the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

select the most accurate machine-learning model as the second machine-learning model.

15. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to:

visually position a graphical indicator proximate to the data value in the graph in response to the first machine-learning model indicating that the data value should be overridden with the override value, wherein the graphical indicator indicates that the data value should be reviewed and is different from the visual directionality cue.

16. The non-transitory computer-readable medium of claim 11, wherein the visual directionality cue is an arrow that is oriented (i) in a first direction to indicate that the override value should be greater than the data value or (ii) in a second direction that is opposite to the first direction to indicate that the override value should be less than the data value.

17. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to:

automatically determine an upper limit and a lower limit for the override value by analyzing the historical data used to generate the time-series projection; and

automatically render visual markers within the GUI indicating the upper limit and the lower limit for the override value.

18. The non-transitory computer-readable medium of claim 17, wherein:

the GUI includes a data table having a column associated with the data value; and

the column includes (i) the data value in a row, (ii) the second input component for inputting the override value in another row, and (iii) the visual directionality cue in yet another row.

19. The non-transitory computer-readable medium of claim 18, wherein the column further includes the upper limit and the lower limit in respective rows.

20. The non-transitory computer-readable medium of claim 17, further comprising program code that is executable by the processing device for causing the processing device to:

analyze the override value and visually display an error indicator within the GUI if the override value is out of a range defined by the upper limit and the lower limit.

21. A method for generating a graphical user interface (GUI) with override guidance, the method comprising:

generating, by a processing device, the GUI, wherein the GUI includes a first input component for receiving a first input indicating a dataset for display within the GUI;

receiving, by the processing device, the first input indicating the dataset via the first input component, wherein the dataset represents a time-series projection spanning a future time period, and wherein the time-series projection was previously generated by analyzing historical data spanning another time period that is prior to the future time period;

in response to receiving the first input, generating, by the processing device, a graph of the time-series projection within the GUI;

automatically feeding, by the processing device, the dataset into a first machine-learning model to obtain an output from the first machine-learning model indicating whether the time-series projection has a data value that should be overridden with an override value;

in response to the output from the first machine-learning model indicating that the time-series projection has the data value that should be overridden;

automatically feeding, by the processing device, the data value as input to a second machine-learning model to obtain an output from the second machine-learning model indicating whether the override value should be greater than or less than the data value; and

automatically rendering, by the processing device, a visual directionality cue within the GUI based on the output from the second machine-learning model, wherein the visual directionality cue is configured to provide guidance about whether the override value should be greater than or less than the data value;

receiving, by the processing device, a second input including the override value via a second input component of the GUI;

in response to receiving the second input, generating, by the processing device, another version of the time-series projection in which the data value is overridden with the override value; and

rendering, by the processing device, the other version of the time-series projection within the GUI.

22. The method of claim 21, further comprising:

generating teaching data using a plurality of time-series projections spanning one or more future time-periods, wherein the teaching data includes (i) predicted values for data points in the plurality of time-series projections, (ii) override values for the data points provided by a particular user, (iii) actual values for the data points gathered during the one or more future time-periods, and (iv) in indication of whether the override values are closer to the actual values than the predicted values.

23. The method of claim 22, further comprising:

training a plurality of machine-learning models using the teaching data to enable the plurality of machine-learning models to determine whether a particular data-value provided to the machine-learning model should be overridden;

determining accuracies of the plurality of machine-learning models using validation data;

comparing the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

selecting the most accurate machine-learning model as the first machine-learning model.

24. The method of claim 22, further comprising:

training a plurality of machine-learning models using the teaching data and outputs from the first machine-learning model;

determining accuracies of the plurality of machine-learning models using validation data;

comparing the accuracies of the plurality of machine-learning models to identify a most accurate machine-learning model among the plurality of machine-learning models; and

selecting the most accurate machine-learning model as the second machine-learning model.

25. The method of claim 21, further comprising:

visually position a graphical indicator proximate to the data value in the graph in response to the first machine-learning model indicating that the data value should be overridden with the override value, wherein the graphical indicator indicates that the data value should be reviewed and is different from the visual directionality cue.

26. The method of claim 21, wherein the visual directionality cue is an arrow that is oriented (i) in a first direction to indicate that the override value should be greater than the data value or (ii) in a second direction that is opposite to the first direction to indicate that the override value should be less than the data value.

27. The method of claim 21, further comprising:

automatically determine an upper limit and a lower limit for the override value by analyzing the historical data used to generate the time-series projection; and

automatically render visual markers within the GUI indicating the upper limit and the lower limit for the override value.

28. The method of claim 27, wherein:

the GUI includes a data table having a column associated with the data value; and

the column includes (i) the data value in a row, (ii) the second input component for inputting the override value in another row, and (iii) the visual directionality cue in yet another row.

29. The method of claim 28, wherein the column further includes the upper limit and the lower limit in respective rows.

30. The method of claim 27, further comprising:

analyzing the override value and visually display an error indicator within the GUI if the override value is out of a range defined by the upper limit and the lower limit.