SYSTEM AND METHOD FOR SCALABLE SEMANTIC STREAM PROCESSING

• US 20120117114A1
• Filed: 11/07/2011
• Published: 05/10/2012
• Est. Priority Date: 11/05/2010
• Status: Active Grant

• Current Assignee: Sirius-Beta Corporation
• Original Assignee: Sirius-Beta Corporation
• Inventors: Goranson, Harold Theodore
• Granted Patent: US 9,117,167 B2
• Time in Patent Office: Days
• US Class Current: 707/776
• CPC Class Codes: G06F 16/41 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06N 5/00 Computing arrangements using knowledge-based models
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A system for collaborative analysis from different processes on different data sources. The system uses a unique approach to lightweight temporary data structures in order to allow communication of interim results among processes, and construction of semantically appropriate reports. The data structures are generated in near real time and their lightweight nature supports massive scaling, including many diverse streaming inputs.

Claims

1. A computer processor system for processing data, comprising:

a communications receiver configured to receive a plurality of streams of data from a plurality of respective sensors;

a recognition processor unit configured to identify features of the streams that match known patterns, and form sb-frames, each sb-frame comprising a segment of data from a stream and semantic information about the segment of data;

a situation processor unit configured to identify situations of interest as control groups;

an engine configured to receive the sb-frames and the control groups as input, and process the sb-frames to identify a presence of situations of interest.

2. The computer system of claim 1, wherein the engine is a wreathing engine and the wreathing engine is configured to synthesize the sb-frames that conform to the control groups to form a series of facts as one or more situated deductions.

3. The computer system of claim 2, wherein the wreathing engine is configured to compare the situated deductions to predetermined threat profiles.

4. The computer system of claim 3, wherein the wreathing engine is configured to trigger an alert if the situated deductions match the predetermined threat profiles.

5. The computer system of claim 1, wherein the semantic data of the sb-frame is data that indicates what the segment of data may represent in a situation.

6. The computer system of claim 1, wherein the streams of data comprise a plurality of media types, and the semantic information of the sb-frames is the same across at least two of the plurality of media types.

7. The computer system of claim 1, wherein the plurality of respective sensors provide data in a plurality of data formats.

8. The computer system of claim 1, wherein the plurality of respective sensors comprises audio and video sensors.

9. The computer system of claim 1, wherein each of the recognition processor unit, the situation processor unit, and the engine are separate computer processors.

10. The computer system of claim 1, wherein the control groups are represented as a mathematical expression.

11. The computer system of claim 1, wherein the streams comprise structured data.

12. The computer system of claim 1, further comprising a computer storage medium storing a plurality of known patterns as recognition processor unit references.

13. The computer system of claim 1, wherein the sb-frames are formed using ontological references that are stored on a computer storage medium.

14. The computer system of claim 1, wherein the situations of interest are stored as mission situation references on a computer storage medium.

15. The computer system of claim 1, comprising a plurality of recognition processor units, each of the plurality of recognition processor units specially designed for a particular media type and/or recognition need.

16. The computer system of claim 1, wherein the recognition processor unit is configured to combine a plurality of sb-frames into a fiber group, wherein the fiber group comprises a plurality of sb-frames related by connectives.

17. The computer system of claim 1, further comprising a router that routes the processed sb-frames to an analysis device.

18. The computer system of claim 1, wherein the plurality of streams of data comprise one or more process flows from each of a plurality of business enterprises, and the situations of interest are products.

19. A method of processing data, comprising:

receiving, using a communications receiver, a plurality of streams of data from a plurality of respective sensors;

identifying features of the streams that match known patterns;

forming sb-frames, each sb-frame comprising a segment of data from a stream and semantic information about the segment of data;

identifying control groups characterizing situations of interest;

synthesizing sb-frames that conform to the control groups to form a series of facts as one or more situated deductions.

20. A device with one or more non-transitory computer-readable mediums storing a program that, when executed, causes a computer processor to:

receive, using a communications receiver, a plurality of streams of data from a plurality of respective sensors;

identify features of the streams that match known patterns;

form sb-frames, each sb-frame comprising a segment of data from a stream and semantic information about the segment of data;

identify control groups characterizing situations of interest;

synthesize sb-frames that conform to the control groups to form a series of facts as one or more situated deductions.