Event detection through text analysis using dynamic self evolving/learning module

• US 9,910,723 B2
• Filed: 01/09/2017
• Issued: 03/06/2018
• Est. Priority Date: 12/02/2013
• Status: Active Grant

• Current Assignee: Finch Computing LLC (Qbase, LLC)
• Original Assignee: Qbase, LLC
• Inventors: Boddhu, Sanjay, Dave, Rakesh
• Primary Examiner(s): Lu, Kuen S
• Application Number:
  US15/402,145
• Publication Number:
  US 20170116054A1
• Time in Patent Office: 421 Days
• Field of Search: 707758
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

A system and method for detecting events based on input data from a plurality of sources. The system may receive input from a plurality of sources containing information about possible events. A method for event detection involves pre-processing and normalizing a data input from a plurality of sources, extracting and disambiguating events and entities, associate event and entities, correlate events and entities associated from a data input to results from a different data source to determine if an event has occurred, and store the detected events in a data storage.

Claims

1. A method comprising:

in response to assigning, by a processor, a first score to a feature identified in a data stream,

determining, by the processor, a second score for the feature via the first score and a number of occurrences of the feature in the data stream, and

identifying, by the processor, an event candidate based on the second score satisfying a first threshold, wherein the event candidate is defined via the feature; and

in response to comparing, by the processor, the feature against a first event model in a first data structure,

determining, by the processor, that the feature does not satisfy the first event model, comparing, by the processor, the feature against a second event model in a second data structure,

storing, by the processor, the feature as a third event model in the second data structure based on the feature not satisfying the second event model, generating, by the processor, a third score for the third event model,

determining, by the processor, a fourth score for the event candidate based on the feature representing an event model that is not in the first data structure and the second data structure,

comparing, by the processor, the third score and the fourth score against a second threshold of the third event model, and

storing, by the processor, the third event model in the first data structure based on the third score and the fourth score being equal to or higher than the second threshold, wherein the first data structure differs from the second data structure based on event model data type.

2. The method of claim 1, wherein the features includes a plurality of features, and further comprising:

identifying, by the processor, a subset of the features according to at least one of the first event model, the second event model, or the third event model; and

determining, by the processor, a score for the subset based on a weight assigned to each of the features in the subset according to the at least one of the first event model, the second event model, or the third event model.

3. The method of claim 2, wherein each of features of the subset satisfies a threshold of the at least one of the first event model, the second event model, or the third event model.

4. The method of claim 1, further comprising:

in response to determining, by the processor, that the feature satisfies the first event model, identifying, by the processor, an event model in the first data structure, wherein the event model is associated with the event candidate.

5. The method of claim 1, wherein the first data structure and the second data structure are distributed among a plurality of servers.

6. A system comprising:

a processor and a memory, wherein the memory stores a set of instructions executable via the processor to;

in response to assigning, by the processor, a first score to a feature identified in a data stream, determining, by the processor, a second score for the feature via the first score and a number of occurrences of the feature in the data stream, and identifying, by the processor, an event candidate based on the second score satisfying a first threshold, wherein the event candidate is defined via the feature;

in response to comparing, by the processor, the feature against a first event model in a first data structure, determining, by the processor, that the feature does not satisfy the first event model, comparing, by the processor, the feature against a second event model in a second data structure, storing, by the processor, the feature as a third event model in the second data structure based on the feature not satisfying the second event model, generating, by the processor, a third score for the third event model, determining, by the processor, a fourth score for the event candidate based on the feature representing an event model that is not in the first data structure and the second data structure, comparing, by the processor, the third score and the fourth score against a second threshold of the third event model, and storing, by the processor, the third event model in the first data structure based on the third score and the fourth score being equal to or higher than the second threshold, wherein the first data structure differs from the second data structure based on event model data type.

7. The system of claim 6, wherein the features includes a plurality of features, and further comprising:

identifying, by the processor, a subset of the features according to at least one of the first event model, the second event model, or the third event model; and

determining, by the processor, a score for the subset based on a weight assigned to each of the features in the subset according to the at least one of the first event model, the second event model, or the third event model.

8. The system of claim 7, wherein each of features of the subset satisfies a threshold of the at least one of the first event model, the second event model, or the third event model.

9. The system of claim 6, wherein the method further comprising:

in response to determining, by the processor, that the feature satisfies the first event model, identifying, by the processor, an event model in the first data structure, wherein the event model is associated with the event candidate.

10. The system of claim 6, wherein the first data structure and the second data structure are distributed among a plurality of servers.