Computer-implemented systems and methods for time series exploration

• US 10,025,753 B2
• Filed: 02/06/2018
• Issued: 07/17/2018
• Est. Priority Date: 07/13/2012
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Leonard, Michael James, Blair, Edward Tilden, Brzezicki, Jerzy Michal, Sglavo, Udo V., Pothireddy, Sujatha
• Primary Examiner(s): TITCOMB, WILLIAM D
• Application Number:
  US15/890,013
• Publication Number:
  US 20180157619A1
• Time in Patent Office: 161 Days
• Field of Search: 715804
• CPC Class Codes: G06F 16/34 Browsing; Visualisation therefor
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and methods are provided for analyzing unstructured time stamped data. A distribution of time-stamped data is analyzed to identify a plurality of potential time series data hierarchies for structuring the data. An analysis of a potential time series data hierarchy may be performed. The analysis of the potential time series data hierarchies may include determining an optimal time series frequency and a data sufficiency metric for each of the potential time series data hierarchies. One of the potential time series data hierarchies may be selected based on a comparison of the data sufficiency metrics. Multiple time series may be derived in a single-read pass according to the selected time series data hierarchy. A time series forecast corresponding to at least one of the derived time series may be generated.

Claims

1. A system comprising:

one or more processors;

one or more computer-readable storage mediums containing instructions configured to cause the one or more processors to perform operations including;

analyzing, in a single-read pass, a distribution of time-stamped unstructured data to identify a plurality of potential time series data hierarchies for structuring the unstructured data, wherein a potential time series data hierarchy is a framework for structuring the unstructured data using time series;

deriving, in the single-read pass, multiple structured time series from the unstructured data according to a selected time series data hierarchy, wherein the selected time series data hierarchy is selected from the plurality of potential time series data hierarchies;

identifying, in the single-read pass, one or more user-specified time series analysis functions from a functions repository based on a function specification and the multiple structured time series, wherein the function specification describes the contents of the functions repository;

generating, in the single-read pass, multiple updated structured time series using the identified time series analysis functions from the functions repository, wherein the updated structured time series is generated by applying the identified time series analysis functions to the multiple structured time series; and

generating an output time series forecast corresponding to one or more of the multiple updated structured time series.

2. The system of claim 1, wherein a potential time series data hierarchy includes a dimension associated with attribute variables, and wherein the function specification maps each function in the functions repository to attribute variables in the time-stamped unstructured data for re-use with other data sets after the time-stamped unstructured data has been structured.

3. The system of claim 2, wherein the function specification uses a function specification API to processes the user-specified time series analysis functions, and to map each function to attribute variables in the time-stamped unstructured data.

4. The system of claim 1, wherein analyzing the distribution of time-stamped unstructured data includes identifying a plurality of time series frequencies, wherein deriving the multiple structured time series from the unstructured data includes using an optimal time series frequency from the plurality of time series frequencies, and wherein the optimal time series frequency is associated with the selected time series data hierarchy.

5. The system of claim 1, wherein generating the output data set includes using an output specification, wherein the output specification describes a form of the output data set.

6. The system of claim 1, wherein the operations further comprise:

generating, in the single-read pass, a set of results using a results specification and the output time series forecast, wherein the results specification describes a tabular and graphical display for visualization of the set of results.

7. The system of claim 1, wherein the set of time series analysis functions includes at least a reduction function, and wherein generating the updated structured time series uses the reduction function.

8. The system of claim 7, wherein the updated structured time series is a summarized, reduced version of the multiple structured time series.

9. The system of claim 1, wherein the set of time series analysis functions includes at least a transformation function, and wherein generating the updated structured time series uses the transformation function.

10. The system of claim 1, wherein the operations further comprise:

generating the output time series forecast using parallelization, wherein each of the multiple updated structured time series are processed separately on separate computational threads.

11. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations including:

analyzing, in a single-read pass, a distribution of time-stamped unstructured data to identify a plurality of potential time series data hierarchies for structuring the unstructured data, wherein a potential time series data hierarchy is a framework for structuring the unstructured data using time series;

deriving, in the single-read pass, multiple structured time series from the unstructured data according to a selected time series data hierarchy, wherein the selected time series data hierarchy is selected from the plurality of potential time series data hierarchies;

identifying, in the single-read pass, one or more user-specified time series analysis functions from a functions repository based on a function specification and the multiple structured time series, wherein the function specification describes the contents of the functions repository;

generating, in the single-read pass, multiple updated structured time series using the identified time series analysis functions from the functions repository, wherein the updated structured time series is generated by applying the identified time series analysis functions to the multiple structured time series; and

generating an output time series forecast corresponding to one or more of the multiple updated structured time series.

12. The computer-program product of claim 11, wherein a potential time series data hierarchy includes a dimension associated with attribute variables, and wherein the function specification maps each function in the functions repository to attribute variables in the time-stamped unstructured data for re-use with other data sets after the time-stamped unstructured data has been structured.

13. The computer-program product of claim 12, wherein the function specification uses a function specification API to processes the user-specified time series analysis functions, and to map each function to attribute variables in the time-stamped unstructured data.

14. The computer-program product of claim 11, wherein analyzing the distribution of time-stamped unstructured data includes identifying a plurality of time series frequencies, wherein deriving the multiple structured time series from the unstructured data includes using an optimal time series frequency from the plurality of time series frequencies, and wherein the optimal time series frequency is associated with the selected time series data hierarchy.

15. The computer-program product of claim 11, wherein generating the output data set includes using an output specification, wherein the output specification describes a form of the output data set.

16. The computer-program product of claim 11, wherein the operations further comprise:

generating, in the single-read pass, a set of results using a results specification and the output time series forecast, wherein the results specification describes a tabular and graphical display for visualization of the set of results.

17. The computer-program product of claim 11, wherein the set of time series analysis functions includes at least a reduction function, and wherein generating the updated structured time series uses the reduction function.

18. The computer-program product of claim 17, wherein the updated structured time series is a summarized, reduced version of the multiple structured time series.

19. The computer-program product of claim 11, wherein the set of time series analysis functions includes at least a transformation function, and wherein generating the updated structured time series uses the transformation function.

20. The computer-program product of claim 11, wherein the operations further comprise:

generating the output time series forecast using parallelization, wherein each of the multiple updated structured time series are processed separately on separate computational threads.

21. A computer-implemented method, the method comprising:

analyzing, in a single-read pass, a distribution of time-stamped unstructured data to identify a plurality of potential time series data hierarchies for structuring the unstructured data, wherein a potential time series data hierarchy is a framework for structuring the unstructured data using time series;

deriving, in the single-read pass, multiple structured time series from the unstructured data according to a selected time series data hierarchy, wherein the selected time series data hierarchy is selected from the plurality of potential time series data hierarchies;

identifying, in the single-read pass, one or more user-specified time series analysis functions from a functions repository based on a function specification and the multiple structured time series, wherein the function specification describes the contents of the functions repository;

generating, in the single-read pass, multiple updated structured time series using the identified time series analysis functions from the functions repository, wherein the updated structured time series is generated by applying the identified time series analysis functions to the multiple structured time series; and

generating an output time series forecast corresponding to one or more of the multiple updated structured time series.

22. The method of claim 21, wherein a potential time series data hierarchy includes a dimension associated with attribute variables, and wherein the function specification maps each function in the functions repository to attribute variables in the time-stamped unstructured data for re-use with other data sets after the time-stamped unstructured data has been structured.

23. The method of claim 22, wherein the function specification uses a function specification API to processes the user-specified time series analysis functions, and to map each function to attribute variables in the time-stamped unstructured data.

24. The method of claim 21, wherein analyzing the distribution of time-stamped unstructured data includes identifying a plurality of time series frequencies, wherein deriving the multiple structured time series from the unstructured data includes using an optimal time series frequency from the plurality of time series frequencies, and wherein the optimal time series frequency is associated with the selected time series data hierarchy.

25. The method of claim 21, wherein generating the output data set includes using an output specification, wherein the output specification describes a form of the output data set.

26. The method of claim 21, further comprising:

generating, in the single-read pass, a set of results using a results specification and the output time series forecast, wherein the results specification describes a tabular and graphical display for visualization of the set of results.

27. The method of claim 21, wherein the set of time series analysis functions includes at least a reduction function, and wherein generating the updated structured time series uses the reduction function.

28. The method of claim 27, wherein the updated structured time series is a summarized, reduced version of the multiple structured time series.

29. The method of claim 21, wherein the set of time series analysis functions includes at least a transformation function, and wherein generating the updated structured time series uses the transformation function.

30. The method of claim 21, further comprising:

generating the output time series forecast using parallelization, wherein each of the multiple updated structured time series are processed separately on separate computational threads.