Identifying cyclic patterns of complex events

• US 9,613,447 B2
• Filed: 02/02/2015
• Issued: 04/04/2017
• Est. Priority Date: 02/02/2015
• Status: Active Grant

• Current Assignee: AirBNB Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Lentz, James L., Van, Jung W., Yheng, Yee Pin
• Primary Examiner(s): CHOW, JEFFREY J
• Application Number:
  US14/611,736
• Publication Number:
  US 20160225171A1
• Time in Patent Office: 792 Days
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 2207/20072 Graph-based image processing

Abstract

Data within a database are displayed to discover patterns of events in time-based data. A first display of a linear timeline indicating events within time-based data is provided. The linear timeline of the first display may be transformed into a second display to cluster the events within the time-based data, where the second display includes one or more from the group of a stacked linear timeline of the events, a polar (annular) timeline of the events, and a helical timeline of the events. Patterns of the events are revealed within the second display to identify event associations.

Claims

1. A method of revealing patterns of events in time-based data comprising:

detecting occurrence of events of a system via a sensor network;

generating on a processing device a first display of a linear timeline indicating the events within time-based data;

transforming the linear timeline of the first display into a second display via the processing device that presents clustering of the events within the time-based data, wherein the second display includes one or more from the group of a stacked linear timeline of the events, a polar timeline of the events, and a helical timeline of the events;

in response to detecting an absence in the second display of patterns of the events to identify event associations contributing to an adverse condition of the system, successively transforming the second display to one or more other timelines of the group via the processing device to reveal the patterns of the events; and

controlling actions of the system via the processing device based on the revealed patterns to address the adverse condition.

2. The method of claim 1, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the stacked linear timeline of the events within a desired time interval, wherein the stacked linear timeline includes a plurality of rows of the events.

3. The method of claim 2, wherein transforming the linear timeline further comprises:

compressing the desired time interval of the stacked linear timeline to enable the events from a row to wrap to a succeeding row.

4. The method of claim 2, wherein transforming the linear timeline further comprises:

compressing the rows of the stacked linear timeline to produce a linear timeline with frequency histograms each indicating occurrence of a quantity of the events.

5. The method of claim 1, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the polar timeline of the events within a desired time interval, wherein the polar timeline includes a generally circular display with frequency histograms each indicating occurrence of a quantity of the events disposed about a circumference of the circular display.

6. The method of claim 5, wherein successively transforming the second display further comprises:

transforming the polar timeline of the events into the helical timeline of the events, wherein the helical timeline of the events includes a generally cylindrical display with events indicated by a spiral path along a cylinder perimeter.

7. The method of claim 6, wherein transforming the polar timeline further comprises:

adjusting a dimension of the cylindrical display to alter a level of granularity of the events within the cylindrical display.

8. The method of claim 1, further comprising:

selecting a cluster of events within the second display and performing an analysis to determine a cause of the cluster of events.

9. A system for detecting patterns of events in time-based data, the system comprising:

a sensor network to detect occurrence of events of within a second system; and

at least one processor configured to;

generate a first display of a linear timeline indicating the events within time-based data;

transform the linear timeline of the first display into a second display that presents clustering of the events within the time-based data, wherein the second display includes one or more from the group of a stacked linear timeline of the events, a polar timeline of the events, and a helical timeline of the events;

in response to detecting an absence in the second display of patterns of the events to identify event associations contributing to an adverse condition of the system, successively transform the second display to one or more other timelines of the group to reveal the patterns of the events; and

control actions of the second system based on the revealed patterns to address the adverse condition.

10. The system of claim 9, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the stacked linear timeline of the events within a desired time interval, wherein the stacked linear timeline includes a plurality of rows of the events.

11. The system of claim 10, wherein transforming the linear timeline further comprises:

compressing the desired time interval of the stacked linear timeline to enable the events from a row to wrap to a succeeding row.

12. The system of claim 10, wherein transforming the linear timeline further comprises:

compressing the rows of the stacked linear timeline to produce a linear timeline with frequency histograms each indicating occurrence of a quantity of the events.

13. The system of claim 9, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the polar timeline of the events within a desired time interval, wherein the polar timeline includes a generally circular display with frequency histograms each indicating occurrence of a quantity of the events disposed about a circumference of the circular display.

14. The system of claim 13, wherein successively transforming the second display further comprises:

transforming the polar timeline of the events into the helical timeline of the events, wherein the helical timeline of the events includes a generally cylindrical display with events indicated by a spiral path along a cylinder perimeter.

15. The system of claim 14, wherein transforming the polar timeline further comprises:

adjusting a dimension of the cylindrical display to alter a level of granularity of the events within the cylindrical display.

16. The system of claim 9, wherein the at least one processor is further configured to:

select a cluster of events within the second display and perform an analysis to determine a cause of the cluster of events.

17. A computer program product for revealing patterns of events in time-based data, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising computer readable program code configured to;

generate a first display of a linear timeline indicating events of a system detected via a sensor network within time-based data;

transform the linear timeline of the first display into a second display that presents clustering of the events within the time-based data, wherein the second display includes one or more from the group of a stacked linear timeline of the events, a polar timeline of the events, and a helical timeline of the events;

in response to detecting an absence in the second display of patterns of the events to identify event associations contributing to an adverse condition of the system, successively transform the second display to one or more other timelines of the group to reveal the patterns of the events; and

control actions of the system based on the revealed patterns to address the adverse condition.

18. The computer program product of claim 17, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the stacked linear timeline of the events within a desired time interval, wherein the stacked linear timeline includes a plurality of rows of the events.

19. The computer program product of claim 17, wherein transforming the linear timeline further comprises:

transforming the linear timeline of the events into the polar timeline of the events within a desired time interval, wherein the polar timeline includes a generally circular display with frequency histograms each indicating occurrence of a quantity of the events disposed about a circumference of the circular display.

20. The computer program product of claim 19, wherein successively transforming the second display further comprises:

transforming the polar timeline of the events into the helical timeline of the events, wherein the helical timeline of the events includes a generally cylindrical display with events indicated by a spiral path along a cylinder perimeter.