DIGITAL PATTERN PROGNOSTICS

• US 20170357828A1
• Filed: 06/05/2017
• Published: 12/14/2017
• Est. Priority Date: 06/10/2016
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: PHILLIPS, Richard Alan
• Granted Patent: US 10,824,761 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Systems and techniques for facilitating digital data prognostics are presented. A system can processes a corpus of stored data, generate respective digital signatures representing respective subsets of the corpus of the stored data, and tag the respective digital signatures with tags corresponding to extrinsic events. The digital signatures can be stored and indexed in a digital signature library. The system can also compare a new digital signature to learned digital signatures in order to identify one or more matches, and prognose an upcoming event associated with the new digital signature based on the matches and generated inferences for the learned digital signatures.

Claims

1. A digital prognostics system, comprising:

a memory having stored thereon computer executable components;

a processor that executes at least the following computer executable components;

a data signature generation component that processes a corpus of stored data and generates respective digital signatures representing respective subsets of the corpus of the stored data, wherein the digital signatures are stored and indexed in a digital signature library;

a tagging component that tags the respective digital signatures with tags corresponding to extrinsic events;

an artificial intelligence component that learns the respective digital signatures and associated tags, and generates inferences regarding the respective digital signatures;

a search component that searches and compares a new digital signature to the learned digital signatures in order to identify one or more matches;

a prognostics component that prognoses an upcoming event associated with the new digital signature based at least in part on the generated inferences and the matches; and

a display component that generates a user interface, for display, that outputs the prognoses in a human interpretable format.

2. The digital prognostics system of claim 1, wherein the data signature generation component generates a digital signature in response to an extrinsic event.

3. The digital prognostics system of claim 1, wherein the data signature generation component generates a digital signature based on a portion of the stored data that is associated with an interval of time before an extrinsic event.

4. The digital prognostics system of claim 1, wherein the data signature generation component generates the new digital signature in response to receiving a portion of the stored data associated with the new digital signature.

5. The digital prognostics system of claim 1, wherein the data signature generation component generates a digital signature based on feedback data received from the user interface.

6. The digital prognostics system of claim 1, wherein the data signature generation component receives the stored data from a database in communication with the digital prognostics system via a network.

7. The digital prognostics system of claim 1, wherein the search component determines that the new digital signature matches a digital signature from the digital signatures stored in the digital signature library.

8. The digital prognostics system of claim 1, wherein the prognostics component triggers an action in response to a determination that the new digital signature matches a digital signature from the learned digital signatures.

9. The digital prognostics system of claim 1, wherein the prognostics component correlates the upcoming event with an asset associated with the upcoming event.

10. A method, comprising:

generating, by a system comprising a processor, a first digital signature based on a portion of stored data;

tagging, by the system, the first digital signature with a tag corresponding to an event extrinsic to the system;

storing, by the system, the first digital signature in a digital signature library;

generating, by the system, inferences regarding the first digital signature;

comparing, by the system, a second digital signature to the first digital signature in order to identify a match;

identifying, by the system, an upcoming event associated with the second digital signature based at least in part on the inferences and the match; and

generating, by the system, a user interface that outputs information associated with the upcoming event in a human interpretable format via a display.

11. The method of claim 10, wherein the generating the first digital signature comprises generating the first digital signature in response to the event.

12. The method of claim 10, wherein the generating the first digital signature comprises determining an interval of time associated with the portion that occurs before the event.

13. The method of claim 10, wherein the generating the first digital signature comprises generating the first digital signature based on input data received via the user interface.

14. The method of claim 10, further comprising generating, by the system, the second digital signature based on another portion of the stored data.

15. The method of claim 10, further comprising receiving, by the system, the stored data from a database in communication with the system via a wireless network.

16. The method of claim 10, further comprising triggering, by the system, an action extrinsic to the system in response to a determination that the second digital signature matches the first digital signature.

17. The method of claim 10, further comprising correlating, by the system, the upcoming event with an asset associated with the portion of the stored data.

18. A computer readable storage device comprising instructions that, in response to execution, cause a system comprising a processor to perform operations, comprising:

generating a first digital fingerprint based on a portion of data stored in a first data store;

generating a tag for the first digital fingerprint to associate the first digital fingerprint with an extrinsic event;

storing the first digital fingerprint in a second data store;

determining inferences associated with the first digital fingerprint;

identifying a match between a second digital fingerprint and the first digital fingerprint, comprising identifying an upcoming event associated with the second digital fingerprint based at least in part on the inferences and the match; and

presenting information associated with the upcoming event in a human interpretable format via a user interface associated with a display.

19. The computer readable storage device of claim 18, wherein the generating the first digital fingerprint comprises generating the first digital fingerprint in response to identifying the extrinsic event.

20. The computer readable storage device of claim 18, wherein the generating the first digital fingerprint comprises determining a period of time associated with the portion that occurs prior to the extrinsic event.