Insight objects as portable user application objects

• US 10,620,790 B2
• Filed: 04/13/2017
• Issued: 04/14/2020
• Est. Priority Date: 11/08/2016
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Ellis, Charles David, Fortier, Dominique, Xue, Liyin, Sharma, Ashvini, Nielsen, Jakob Peter, Otero, Carlos Augusto, Bhojawala, Safiya, Neylan, Catherine William
• Primary Examiner(s): Ries, Laurie A
• Application Number:
  US15/487,066
• Publication Number:
  US 20180129373A1
• Time in Patent Office: 1,097 Days
• Field of Search: 715215
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 40/18 of spreadsheets form-filling G06F40/174
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: G06T 2200/24 involving graphical user interfaces [GUIs]

Abstract

Systems, methods, and software for data visualization frameworks are provided herein. An exemplary method includes identifying a target dataset from which to determine data insights for presentation in an insight interface to the user application. The method includes determining data insight candidates for the target dataset based at least on usage modalities associated with processing one or more past datasets, and establishing content of the data insight candidates according to at least the target dataset and the usage modalities, where the content of each of the data insight candidates includes at least one insight object described by object metadata that indicates at least a processing lineage used to produce the at least one insight object.

Claims

1. A method of providing a data visualization framework to a user application, the method comprising:

identifying a target dataset from which to determine one or more data insights for presentation in an insight interface to the user application;

determining data insight candidates for the target dataset based at least on usage modalities associated with processing one or more past datasets;

establishing content of the data insight candidates according to at least the target dataset and the usage modalities, wherein the content of each of the data insight candidates includes at least one insight object described by object metadata that indicates at least a displayable processing lineage used to produce the at least one insight object, wherein the processing lineage comprises an intermediate step used in processing the target dataset into the at least one insight object;

displaying the at least one insight object with the displayable processing lineage used to produce it;

displaying a plurality of selectable target questions related to the displayed insight object;

receiving a selection of one of the target questions; and

displaying a rationale detailing a reason that the displayed insight object was chosen in response to the selection.

2. The method of claim 1, further comprising:

based at least on receiving a user selection of a first data insight candidate among a listing of the data insight candidates presented in the insight interface, producing the first data insight candidate for inclusion in a user interface to the user application.

3. The method of claim 1, further comprising:

presenting the data insight candidates in a ranked order determined according to at least a relevance based on the usage modalities.

4. The method of claim 1, wherein the content further includes data analysis conclusion verbiage that textually summarizes the at least one data insight object;

and further comprising;

determining the data analysis conclusion verbiage for each of the data insight candidates based at least on analyzing the target dataset for data patterns in the target dataset and selecting the data analysis conclusion verbiage to embody the data patterns to the user.

5. The method of claim 1, wherein the object metadata that indicates the processing lineage comprises a formula employed to generate the at least one insight object.

6. The method of claim 1, wherein the object metadata is configured to accompany the at least one insight object when the at least one insight object is embedded into a document associated with another user application than the user application in which the target dataset is identified.

7. The method of claim 1, further comprising:

based on a user indication selecting a chain of inference representation of the first data insight candidate, producing the chain of inference representation including one or more graphs, charts, or tables based at least on the target dataset and the object metadata associated with the first data insight candidate;

wherein the chain of inference representation comprises a visualization of one or more intermediate steps used in processing the target dataset into an insight object of the first data insight candidate.

8. The method of claim 1, wherein the usage modalities comprise indications of past application usage patterns derived from one or more of past data analyses, organizational analysis preferences, user data files, and user application telemetry records.

9. A data visualization framework, comprising:

one or more computer readable hardware storage device;

a processing system operatively coupled with the one or more computer readable storage device; and

a data insight service comprising program instructions stored on the one or more computer readable hardware storage device that, based on being read and executed by the processing system, direct the processing system to at least;

identify a target dataset from which to determine one or more data visualizations for presentation in an insight interface to a user application;

determine data insight candidates for the target dataset based at least on usage modalities associated with processing one or more past datasets;

establish content of the data insight candidates according to at least the target dataset and the usage modalities, wherein the content of each of the data insight candidates includes at least one insight object described by object metadata that indicates at least a displayable processing lineage used to produce the at least one insight object, wherein the processing lineage comprises an intermediate step used in processing the target dataset into the at least one insight object;

display the at least one insight object with the displayable processing lineage used to produce it;

display a plurality of selectable target questions related to the displayed insight object;

receiving a selection of one of the target questions; and

displaying a rationale detailing a reason that the displayed insight object was chosen in response to the selection.

10. The data visualization framework of claim 9, comprising further program instructions, based on being executed by the processing system, direct the processing system to at least:

based at least on receiving a user selection of a first data insight candidate among a listing of the data insight candidates presented in the insight interface, produce the first data insight candidate for inclusion in a user interface to the user application.

11. The data visualization framework of claim 9, comprising further program instructions, based on being executed by the processing system, direct the processing system to at least:

present the data insight candidates in a ranked order determined according to at least a relevance based on the usage modalities.

12. The data visualization framework of claim 9, wherein the content further includes data analysis conclusion verbiage that textually summarizes the at least one insight object;

and comprising further program instructions, based on being executed by the processing system, direct the processing system to at least;

determine the data analysis conclusion verbiage for each of the data insight candidates based at least on analyzing the target dataset for data patterns in the target dataset; and

select the data analysis conclusion verbiage to embody the data patterns to the user.

13. The data visualization framework of claim 9, wherein the object metadata that indicates the processing lineage comprises a formula employed to generate the at least one insight object.

14. The data visualization framework of claim 9, wherein the object metadata is configured to accompany the at least one insight object when the at least one insight object is embedded into a document associated with another user application than the user application in which the target dataset is identified.

15. The data visualization framework of claim 9, comprising further program instructions, based on being executed by the processing system, direct the processing system to at least:

based on a user indication selecting a chain of inference representation of the first data insight candidate, produce the chain of inference representation including one or more graphs, charts, or tables based at least on the target dataset and the object metadata associated with the first data insight candidate;

wherein the chain of inference representation comprises a visualization of intermediate steps used in processing the target dataset into an insight object of the first data insight candidate.

16. The data visualization framework of claim 9, wherein the usage modalities comprise indications of past application usage patterns derived from one or more of past data analyses, organizational analysis preferences, user data files, and user application telemetry records.

17. A method of operating a user application platform, comprising:

presenting a user interface to a spreadsheet application comprising a workbook portion from which a user identifies a target dataset for a data insight service;

providing the target dataset to the data insight service and responsively receiving one or more candidate data insights for at least a portion of the target dataset that are determined based at least on data analysis preferences established based in part on past usage modalities of the user;

presenting an insight user interface to the spreadsheet application listing the candidate data insights;

responsive to the user identifying a selected candidate data insight, inserting the selected candidate data insight into the workbook portion and associating the selected candidate data insight with metadata representative of a displayable processing lineage used to produce the selected candidate data insight, wherein the processing lineage comprises an intermediate step used in processing the target dataset into the selected candidate data insight;

presenting the candidate data insight with the displayable processing lineage used to produce it;

presenting a plurality of selectable target questions related to the candidate data insight;

receiving a selection of one of the target questions; and

presenting a rationale detailing a reason that the candidate data insight was chosen in response to the selection.

18. The method of claim 17, further comprising:

presenting the candidate data insights in a ranked order according to at least a relevance determined from the data analysis preferences.

19. The method of claim 17, wherein the metadata comprises a formula employed to generate the at least one insight object.

20. The method of claim 17, further comprising:

based on a user indication selecting a chain of inference representation of the selected candidate data insight, producing the chain of inference representation including one or more graphs, charts, or tables based at least on the target dataset and the metadata associated with the selected candidate data insight;

wherein the chain of inference representation comprises a visualization of intermediate steps used in processing the target dataset into an insight object of the selected candidate data insight.