HIGH-LEVEL SPECIALIZATION LANGUAGE FOR SCALABLE SPATIOTEMPORAL PROBABILISTIC MODELS

• US 20140365404A1
• Filed: 06/11/2013
• Published: 12/11/2014
• Est. Priority Date: 06/11/2013
• Status: Abandoned Application

• Current Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Original Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Inventors: Bart, Evgeniy, Price, Robert R., Bobrow, Daniel G.
• Application Number:
  US13/915,551
• Publication Number:
  US 20140365404A1
• Time in Patent Office: Days
• US Class Current: 706/11
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

One embodiment of the present invention provides a system for clustering heterogeneous events using user-provided constraints. During operation, the system estimates, based on a probabilistic model, a distribution of events across clusters such that each cluster includes a set of events. Next, the system estimates a probability distribution for an event property associated with each cluster. The system receives heterogeneous event data, and analyzes the heterogeneous event data to determine the probability distribution of event properties of clusters and to assign events to clusters. The system receives user input specifying the user-provided constraints for specializing the probabilistic model, and performs at least one of: re-computing the assignment of events to clusters, and re-determining the probability distribution of event properties of clusters based on the user input.

Claims

1. A computer-executable method performed by a system for clustering heterogeneous events using user-provided constraints, comprising:

estimating, based on a probabilistic model, a distribution of events across clusters such that each cluster includes a set of events;

estimating a probability distribution for an event property associated with each cluster;

receiving heterogeneous event data;

analyzing the heterogeneous event data to determine the probability distribution of event properties of clusters and to assign events to clusters;

receiving user input specifying the user-provided constraints for specializing the probabilistic model; and

performing at least one of;

re-computing the assignment of events to clusters; and

re-determining the probability distribution of event properties of clusters based on the user input.

2. The method of claim 1, wherein the user-provided constraints specify that two or more events belong to the same cluster in the probabilistic model.

3. The method of claim 1, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed according to the probabilistic model, and that events associated with time after the particular time are processed according to another probabilistic model.

4. The method of claim 1, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed as a first event type, and that events associated with time after the particular time are processed as a second event type.

5. The method of claim 1, wherein the user-provided constraints specify relationships between variables in the probabilistic model.

6. The method of claim 1, further comprising receiving user input that specifies parameters associated with events are same when locations associated with the events are the same.

7. A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for clustering heterogeneous events using user-provided constraints, the method comprising:

estimating, based on a probabilistic model, a distribution of events across clusters such that each cluster includes a set of events;

estimating a probability distribution for an event property associated with each cluster;

receiving heterogeneous event data;

analyzing the heterogeneous event data to determine the probability distribution of event properties of clusters and to assign events to clusters;

receiving user input specifying the user-provided constraints for specializing the probabilistic model; and

performing at least one of;

re-computing the assignment of events to clusters; and

re-determining the probability distribution of event properties of clusters based on the user input.

8. The computer-readable storage medium of claim 7, wherein the user-provided constraints specify that two or more events belong to the same cluster in the probabilistic model.

9. The computer-readable storage medium of claim 7, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed according to the probabilistic model, and that events associated with time after the particular time are processed according to another probabilistic model.

10. The computer-readable storage medium of claim 7, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed as a first event type, and that events associated with time after the particular time are processed as a second event type.

11. The computer-readable storage medium of claim 7, wherein the user-provided constraints specify relationships between variables in the probabilistic model.

12. The computer-readable storage medium of claim 7, wherein the computer-readable storage medium stores additional instructions that, when executed, cause the computer to perform additional steps comprising:

receiving user input that specifies parameters associated with events are same when locations associated with the events are the same.

13. A computing system for clustering heterogeneous events using user-provided constraints, the system comprising:

one or more processors,

a computer-readable medium coupled to the one or more processors having instructions stored thereon that, when executed by the one or more processors, cause the one or more processors to perform operations comprising;

estimating, based on a probabilistic model, a distribution of events across clusters such that each cluster includes a set of events;

estimating a probability distribution for an event property associated with each cluster;

receiving heterogeneous event data;

analyzing the heterogeneous event data to determine the probability distribution of event properties of clusters and to assign events to clusters;

receiving user input specifying the user-provided constraints for specializing the probabilistic model; and

performing at least one of;

re-computing the assignment of events to clusters; and

re-determining the probability distribution of event properties of clusters based on the user input.

14. The computing system of claim 13, wherein the user-provided constraints specify that two or more events belong to the same cluster in the probabilistic model.

15. The computing system of claim 13, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed according to the probabilistic model, and that events associated with time after the particular time are processed according to another probabilistic model.

16. The computing system of claim 13, wherein the user-provided constraints specify that events associated with time prior to a particular time are processed as a first event type, and that events associated with time after the particular time are processed as a second event type.

17. The computing system of claim 13, wherein the user-provided constraints specify relationships between variables in the probabilistic model.

18. The computing system of claim 13, further comprising receiving user input that specifies parameters associated with events are same when locations associated with the events are the same.