Automatic time series exploration for business intelligence analytics

• US 10,453,007 B2
• Filed: 05/18/2015
• Issued: 10/22/2019
• Est. Priority Date: 05/18/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chu, Yea Jane, Han, Sier, Shyr, Jing-Yun
• Primary Examiner(s): Feacher, Renae
• Assistant Examiner(s): Leal, Hector
• Application Number:
  US14/715,378
• Publication Number:
  US 20160342909A1
• Time in Patent Office: 1,618 Days
• Field of Search: 705 711
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs

Abstract

Techniques are described for generating characterizations of time series data. In one example, a method includes extracting a trend-cycle component, a seasonal component, and an irregular component from a time series of data. The method further includes performing one or more pattern analyzes on the trend-cycle component, the seasonal component, and the irregular component. The method further includes, for each pattern analysis of the one or more pattern analyzes, performing a comparison of an analytic result of the respective pattern analysis to a selected significance threshold for the respective pattern analysis to determine if the analytic result passes the significance threshold for the respective pattern analysis. The method further includes generating an output for each of the analytic results that pass the significance threshold for the respective pattern analysis.

Claims

1. A computer system comprising:

one or more processors, one or more computer-readable memories, and one or more computer-readable tangible storage devices;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to retrieve, from one or more databases, a time series of data in response to receiving a request from a client computing device;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to decompose the time series of data to extract a trend-cycle component, a seasonal component, and an irregular component from the time series of data;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to distribute the trend-cycle component, the seasonal component, and the irregular component of the time series of data into different mappers of a distributed computing system;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to perform, by the different mappers of the distributed computing system, and at least partially in parallel, different pattern analyses on the trend-cycle component, the seasonal component, and the irregular component, wherein performing the different pattern analyses comprises;

performing, on the trend-cycle component, at least one of a turning point detection analysis or an overall trend analysis using a time correlation statistic;

performing, on the seasonal component, at least one of a seasonal pattern significance test or an unusual season detection analysis; and

performing, on the irregular component, at least one of an outlier detection analysis, a large variance interval detection analysis, or an autocorrelation function analysis;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, and for each of the different pattern analyses on one of the trend-cycle component, the seasonal component, and the irregular component, to compare a respective analytic result of the respective pattern analysis to a respective significance threshold associated with the respective pattern analysis to determine whether the respective analytic result passes the respective significance threshold;

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to generate one or more data visualizations that are selected for display, wherein the selected one or more data visualizations include each respective analytic result that passes the respective significance threshold associated with the respective pattern analysis; and

program instructions, stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, to output, for display at the client computing device, the one or more data visualizations, wherein outputting the one or more data visualizations includes outputting at least one data visualization that includes the respective analytic result of the irregular component displayed in relation to a combined display of the respective analytic results of both the trend-cycle and seasonal components within the at least one visualization.

2. The computer system of claim 1,

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of the time correlation statistic to a selected time correlation significance threshold, and program instructions to compare a result of the turning point detection analysis to a selected turning point significance threshold.

3. The computer system of claim 1,

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of a p-value of the seasonal pattern significance test to an alpha significance threshold.

4. The computer system of claim 1,

wherein the program instructions to perform the seasonal pattern significance test on the seasonal component comprise program instructions to determine an effect size value to indicate a strength of the seasonal component, and

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of the effect size value to one or more selected effect size significance thresholds.

5. The computer system of claim 1,

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of z-scores of seasons to a selected positive or negative z-score significance threshold.

6. The computer system of claim 1,

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise at least one of program instructions to compare a result of z-score values from the outlier detection analysis to a selected z-score significance threshold, program instructions to compare a p-value of a large variance interval test statistic from the large variance interval detection analysis to a selected alpha significance threshold, or program instructions to compare a result of the autocorrelation function analysis to a selected confidence level significance threshold.

7. A computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors to:

responsive to receiving a request from a client computing device, retrieve, from one or more databases, a time series of data;

decompose the time series of data to extract a trend-cycle component, a seasonal component, and an irregular component from the time series of data;

distribute the trend-cycle component, the seasonal component, and the irregular component of the time series of data into different mappers of the distributed computing system;

perform, by the different mappers of the distributed computing system, and at least partially in parallel, different pattern analyses on the trend-cycle component, the seasonal component, and the irregular component, wherein performing the different pattern analyses comprises;

performing, on the trend-cycle component, at least one of a turning point detection analysis or an overall trend analysis using a time correlation statistic;

performing, on the seasonal component, at least one of a seasonal pattern significance test or an unusual season detection analysis; and

performing, on the irregular component, at least one of an outlier detection analysis, a large variance interval detection analysis, or an autocorrelation function analysis;

for each of the different pattern analyses on the trend-cycle component, the seasonal component, and the irregular component, compare a respective analytic result of the respective pattern analysis to a respective significance threshold associated with the respective pattern analysis to determine whether the respective analytic result passes the respective significance threshold;

generate one or more data visualizations that are selected for display, wherein the selected one or more data visualizations include each respective analytic result that passes the respective significance threshold associated with the respective pattern analysis; and

output, for display at the client computing device, the one or more data visualizations, wherein outputting the one or more data visualizations includes outputting at least one data visualization that includes the respective analytic result of the irregular component displayed in relation to a combined display of the respective analytic results of both the trend-cycle and seasonal components within the at least one visualization.

8. The computer program product of claim 7,

wherein the program instructions to compare the respective analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of the time correlation statistic to a selected time correlation significance threshold, and program instructions to compare a result of the turning point detection analysis to a selected turning point significance threshold.

9. The computer program product of claim 7,

wherein the program instructions to perform the seasonal pattern significance test on the seasonal component comprise program instructions to determine an effect size value to indicate a strength of the seasonal component, and

wherein the program instructions to perform the comparison of the analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise program instructions to compare a result of a p-value of the seasonal pattern significance test to an alpha significance threshold, program instructions to compare a result of the effect size value to one or more selected effect size significance thresholds, and program instructions to compare a result of z-scores of seasons to a selected positive or negative z-score significance threshold.

10. The computer program product of claim 7,

wherein the program instructions to perform the comparison of the analytic result of the respective pattern analysis to the respective significance threshold associated with the respective pattern analysis comprise at least one of program instructions to compare a result of z-score values from the outlier detection analysis to a selected z-score significance threshold, program instructions to compare a p-value of a large variance interval test statistic from the large variance interval detection analysis to a selected alpha significance threshold, or program instructions to compare a result of the autocorrelation function analysis to a selected confidence level significance threshold.