Complex event evaluation systems and methods

• US 20080114574A1
• Filed: 12/20/2006
• Published: 05/15/2008
• Est. Priority Date: 11/09/2006
• Status: Active Grant

• Current Assignee: Institute For Information Industry
• Original Assignee: Institute For Information Industry
• Inventors: Li, Jen-Feng, Huang, Wen-Ju, Hsu, Chih-Hao, Chen, Pin-Chan
• Granted Patent: US 7,415,391 B2
• Time in Patent Office: Days
• US Class Current: 703/2
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

Complex event evaluation systems and methods are disclosed. A mathematics model for complex events and corresponding factors, and calculates a critical level of respective complex events according to weightings of respective factors in the mathematics model are provided. If several complex events occur simultaneously, a priority of respective complex events is calculated according to the historic occurrence of respective complex events. The critical level and priority of respective complex events are provided to users for decision making. Additionally, the mathematics model is established and adjusted according to the feedback scores for complex events.

Claims

1. A complex event evaluation system, comprising:

an event aggregation module receiving a plurality of first events, each comprising at least one event value, and generating an event pool comprising at least one second event according to the first events;

a weight calculation model comprising a weight calculation function corresponding to the at least one second event, where the weight calculation function comprises weightings corresponding to respective first events; and

a weight score calculation module calculating a weight score of the at least one second event according to the event values corresponding to respective first events and the weight calculation function corresponding to the at least one second event, evaluating a critical level of the at least one second event according to the weight score of the at least one second event retrieving the weight scores of the historic events of the at least one second events from the event history database, normalizing the weight score of the at least one second event accordingly to obtain a normalized weight score of the at least one second event, and comparing the normalized weight score with at least one predefined threshold to obtain the critical level of the at least one second event.

5. The system of claim 1 further comprising an event filtering module comparing the first events with at least one filtering rule, and outputting the first events to the event aggregation module if the first events conform to the filtering rule.

6. The system of claim 1 wherein the weight calculation model comprises at least one regression model.

2. (canceled)

3. The system of claim 2 wherein the weight score calculation module further determines a light according to the critical level of the at least one second event, and displays the light in a graphic interface.

4. The system of claim 2 further comprising a priority evaluation module retrieving historic occurrence probability of the at least one second event under the occurrence of the event pool from the event history database, calculating a priority of the at least one second event according to the historic occurrence probability and the normalized weight score of the at least one second event, and determines a processing order for the at least one second event according to the priority thereof.

7. A complex event evaluation method, comprising:

receiving a plurality of first events, each comprising at least one event value;

generating an event pool comprising at least one second event according to the first events;

providing a weight calculation model comprising a weight calculation function corresponding to the at least one second event, where the weight calculation function comprises weightings corresponding to respective first events;

calculating a weight score of the at least one second event according to the event values corresponding to respective first events and the weight calculation function corresponding to the at least one second event;

evaluating a critical level of the at least one second event according to the weight score of the at least one second event;

retrieving the weight scores of the historic events of the at least one second events from an event history database;

normalizing the weight score of the at least one second event accordingly to obtain a normalized weight score of the at least one second event; and

comparing the normalized weight score with at least one predefined threshold to obtain the critical level of the at least one second event.

11. The method of claim 7 further comprising:

comparing the first events with at least one filtering rule; and

outputting the first events to the event aggregation module if the first events conform to the filtering rule.

12. The method of claim 7 wherein the weight calculation model comprises at least one regression model.

8. (canceled)

9. The method of claim 8 further comprising:

determining a light according to the critical level of the at least one second event; and

displaying the light in a graphic interface.

10. The method of claim 8 further comprising:

retrieving historic occurrence probability of the at least one second event under the occurrence of the event pool from the event history database;

calculating a priority of the at least one second event according to the historic occurrence probability and the normalized weight score of the at least one second event; and

determines a processing order for the at least one second event according to the priority thereof.

13. A machine-readable storage medium comprising a computer program, which, when executed, causes a device to perform a complex event evaluation method, the method comprising:

receiving a plurality of first events, each comprising at least one event value;

generating an event pool comprising at least one second event according to the first events;

providing a weight calculation model comprising a weight calculation function corresponding to the at least one second event, where the weight calculation function comprises weightings corresponding to respective first events;

calculating a weight score of the at least one second event according to the event values corresponding to respective first events and the weight calculation function corresponding to the at least one second event; and

evaluating a critical level of the at least one second event according to the weight score of the at least one second event;

retrieving the weight scores of the historic events of the at least one second events from an event history database;

normalizing the weight score of the at least one second event accordingly to obtain a normalized weight score of the at least one second event; and

comparing the normalized weight score with at least one predefined threshold to obtain the critical level of the at least one second event.

17. The storage medium of claim 13 wherein the method further comprises:

comparing the first events with at least one filtering rule; and

outputting the first events to the event aggregation module if the first events conform to the filtering rule.

18. The storage medium of claim 13 wherein the weight calculation model comprises at least one regression model.

14. (canceled)

15. The storage medium of claim 14 wherein the method further comprises:

determining a light according to the critical level of the at least one second event; and

displaying the light in a graphic interface.

16. The storage medium of claim 14 wherein the method further comprises:

retrieving historic occurrence probability of the at least one second event under the occurrence of the event pool from the event history database;

calculating a priority of the at least one second event according to the historic occurrence probability and the normalized weight score of the at least one second event; and

determines a processing order for the at least one second event according to the priority thereof.