System and method for generation of a heuristic

Patent metadata

Field	Value
Publication number	US-9818060-B2
Application number	US-201615354122-A
Country	US
Kind code	B2
Filing date	Nov 17, 2016
Priority date	Feb 6, 2014
Publication date	Nov 14, 2017
Grant date	Nov 14, 2017

Abstract

A system and method for generating a heuristic is provided. A heuristic is capable of identifying data patterns. The method includes: extracting a data set from multiple input sources; creating a set of unique elements used across the data set; organizing the data set into a geometric structure; grouping portions of the data in the geometric structure into a plurality sub geometric structures; determining base attributes for each sub geometric structure using the set of unique elements; identifying trends in the base attributes among the sub geometric structures; and outputting the heuristic as a combination of the base attributes and the trends.

First claim

What is claimed is:

1. A method for monitoring a computing system by cognitive fingerprinting, comprising:
extracting, by a processor, a data set from a plurality of input sources across a monitored computing system;
compiling a dictionary of unique elements identified in the data set, each unique element representing a unique discrete portion of the data set corresponding to a portion of the monitored computing system or a status of the portion of the monitored computing system;
organizing the unique elements of the data set into a two-dimensional geometric structure, portions of the data in the geometric structure further grouped into a plurality of sub geometric structures;
generating a heuristic describing the two or greater-dimensional geometric structure, the plurality of sub geometric structures, one or more base properties determined for each sub geometric structure, the base properties determined using the dictionary of unique elements, and one or more trends identified in the base properties among the sub geometric structures;
applying one or more pattern recognizers to the heuristic to produce a set of fingerprint elements corresponding to one or more discrete qualities in or attributes of the data set corresponding to the monitored computing system;
creating a cognitive fingerprint corresponding to the fingerprint elements;
comparing, by the processor, the cognitive fingerprint to a previously recorded cognitive fingerprint by taking a sum of squared differences between each element in the final set of elements and each element of the previously recorded cognitive fingerprint to determine a scalar value for comparison to a threshold of a target goal; and
one or more of shutting down the monitored computing system or quarantining at least a portion of the extracted data in response to the scalar value being within the threshold of the target goal.

2. The method of claim 1 , further comprising:
assigning a magnitude value to each element in the set of unique elements.

3. The computer implemented method of claim 2 , wherein one of the base properties determined is a presence attribute.

4. The method of claim 2 , wherein one of the base properties determined is a magnitude attribute.

5. The method of claim 2 , wherein one of the base properties determined is a position attribute.

6. The method of claim 2 , wherein outputs of all base properties are normalized amongst each other.

7. The method of claim 2 , wherein the trends are identified by using a linear progression function.

8. The method of claim 2 , further comprising:
generating a heuristic definition including references to the sub geometric structures.

9. A computer monitoring system, comprising:
a processor; and
memory coupled to the processor and storing instructions that, when executed by the processor, cause the computer monitoring system to:
extract a data set from a plurality of input sources across a monitored computer system;
compile a dictionary of unique elements identified in the data set, each unique element representing a unique discrete portion of the data set corresponding to a portion of the monitored computer system or a status of the portion of the monitored computer system;
organize the unique elements of the data set into a two-dimensional geometric structure, portions of the data in the geometric structure further grouped into a plurality of sub geometric structures;
generate a heuristic describing the two or greater-dimensional geometric structure, the plurality of sub geometric structures, one or more base properties determined for each sub geometric structure, the base properties determined using the dictionary of unique elements, and one or more trends identified in the base properties among the sub geometric structures;
apply one or more pattern recognizers to the heuristic to produce a set of fingerprint elements corresponding to one or more discrete qualities in or attributes of the data set corresponding to the monitored computer system;
create a cognitive fingerprint corresponding to the fingerprint elements;
compare the cognitive fingerprint to a previously recorded cognitive fingerprint by taking a sum of squared differences between each element in the final set of elements and each element of the previously recorded cognitive fingerprint to determine a scalar value for comparison to a threshold of a target goal; and
one or more of shutting down the monitored computer system or quarantining at least a portion of the extracted data in response to the scalar value being within the threshold of the target goal.

10. The computer system of claim 9 , wherein the memory further stores instructions to cause the processor to assign a magnitude value to each element in the set of unique elements.

11. The computer system of claim 10 , wherein one of the base properties determined is a presence attribute.

12. The computer system of claim 10 , wherein one of the base properties determined is a magnitude attribute.

13. The computer system of claim 10 , wherein one of the base properties determined is a position attribute.

14. The computer system of claim 10 , wherein the memory further stores instructions to cause the processor to normalize outputs of all base properties amongst each other.

15. The computer system of claim 10 , wherein the memory further stores instructions to cause the processor to identify the trends using a linear progression function.

16. The computer system of claim 10 , wherein the memory further stores instructions to cause the processor to generate a heuristic definition including references to the sub geometric structures.

17. A method for monitoring a motor by cognitive fingerprinting, comprising:
extracting, by a processor, a data set from a plurality of input sources across a monitored motor;
compiling a dictionary of unique elements identified in the data set, each unique element representing a unique discrete portion of the data set corresponding to a portion of the monitored engine or a status of the portion of the monitored motor;
organizing the unique elements of the data set into a two-dimensional geometric structure, portions of the data in the geometric structure further grouped into a plurality of sub geometric structures;
generating a heuristic describing the two or greater-dimensional geometric structure, the plurality of sub geometric structures, one or more base properties determined for each sub geometric structure, the base properties determined using the dictionary of unique elements, and one or more trends identified in the base properties among the sub geometric structures;
applying one or more pattern recognizers to the heuristic to produce a set of fingerprint elements corresponding to one or more discrete qualities in or attributes of the data set corresponding to the monitored motor;
creating a cognitive fingerprint corresponding to the fingerprint elements;
comparing, by the processor, the cognitive fingerprint to a previously recorded cognitive fingerprint by taking a sum of squared differences between each element in the final set of elements and each element of the previously recorded cognitive fingerprint to determine a scalar value for comparison to a threshold of a target goal; and
one or more of shutting down the monitored motor or adjusting an operational parameter of the monitored motor in response to the scalar value being within the threshold of the target goal.

18. The method of claim 17 , wherein the motor is one of an electric motor, an internal combustion engine, a diesel

Assignees

• Sparkcognition Inc

Inventors

• Husain Syed Mohammad Amir

Classifications

• G06N5/01 Primary

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N5/025 Primary

  Extracting rules from data · CPC title

• G06N5/047

  Pattern matching networks; Rete networks · CPC title

• G06N5/003 Primary

  Physics · mapped topic