Granted Patent B2

US 8,341,110 · App. 13/073,327 · Granted Dec 25, 2012

Abstract

Claims (74)

1. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature, wherein the temporal feature is designated based upon a correlation between the temporal feature and the event type;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

2. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein the step of determining an intensity function includes measuring a time period between each occurrence of multiple instances of the event type and the at least one temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

3. The method of claim 2 wherein the step of measuring is performed based upon at least one past instance of the event type.

4. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature, wherein the at least one temporal feature has before and after conditions;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature, and wherein the time window break is determined such that the at least one temporal feature's before and after conditions are at maximum separation; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

5. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein the intensity function is a probability density function;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

6. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein the step of determining an intensity function is associated with a geospatial area of concern;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

7. A method of forecasting a critical time point for an event type, comprising the steps of:

causing at least one processor to execute a plurality of instructions stored in at least one memory device to designate at least one temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine an intensity function representing an expected value of event type instances corresponding to the temporal feature;

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to determine a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature, wherein the time window break is determined so as to maximize the difference between expected value of event type instances in a first time series, and expected value of event type instances in a second time series; and

causing the at least one processor to execute a plurality of instructions stored in the at least one memory device to assess whether the time window break is a critical time point for the event type.

8. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature, wherein the temporal feature is designated based upon a correlation between the temporal feature and the event type;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

assessing whether the time window break is a critical time point for the event type.

9. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein determining an intensity function includes measuring a time period between each occurrence of multiple instances of the event type and the at least one temporal feature;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

assessing whether the time window break is a critical time point for the event type.

10. The system of claim 9 wherein the step of measuring is performed based upon at least one past instance of the event type.

11. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature, wherein the at least one temporal feature has before and after conditions;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature, and wherein the time window break is determined such that the at least one temporal feature's before and after conditions are at maximum separation; and

assessing whether the time window break is a critical time point for the event type.

12. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein the intensity function is a probability density function;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

assessing whether the time window break is a critical time point for the event type.

13. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature, wherein the step of determining an intensity function is associated with a geospatial area of concern;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature; and

assessing whether the time window break is a critical time point for the event type.

14. A system for forecasting a critical time point for an event type, comprising:

a processor; and

a non-transitory computer-readable medium including instructions that, when executed by the processor perform a method, the method comprising:

designating at least one temporal feature;

determining an intensity function representing an expected value of event type instances corresponding to the temporal feature;

determining a time window break associated with the intensity function, the window break representing observed change across a time threshold for the temporal feature, wherein the time window break is determined so as to maximize the difference between expected value of event type instances in a first time series, and expected value of event type instances in a second time series; and

assessing whether the time window break is a critical time point for the event type.

Assignments (9)