Systems and methods for diagnosing problems from error logs using natural language processing

Patent metadata

Field	Value
Publication number	US-11568134-B2
Application number	US-202016989736-A
Country	US
Kind code	B2
Filing date	Aug 10, 2020
Priority date	Jun 30, 2017
Publication date	Jan 31, 2023
Grant date	Jan 31, 2023

Abstract

Disclosed is a solution for diagnosing problems from logs used in an application development environment. A random sample of log statements is collected. The log statements can be completely unstructured and/or do not conform to any natural language. The log statements are tagged with predefined classifications. A natural language processing (NLP) classifier model is trained utilizing the log statements tagged with the predefined classification. New log statements can be classified into the plurality of predefined classifications utilizing the trained NLP classifier model. From the log statements thus classified, statements having a problem classification can be identified and presented through a dashboard running in a browser. Outputs from the trained NLP classifier model can be provided as input to another trained model for automatically and quickly identifying a type of problem associated with the statements, eliminating a need to manually sift through tens or hundreds of thousands of lines of logs.

First claim

What is claimed is:

1. A system for diagnosing problems from logs, the system comprising:
a processor;
a non-transitory computer-readable medium comprising instructions for:
receiving new log statements;
classifying the new log statements into at least one of a plurality of classifications based on an NLP feature associated with the log statements utilizing a NLP classifier model, wherein the plurality of predefined classifications include a problem classification and wherein the NLP classifier model was trained by:
utilizing log statements tagged with the at least one of the plurality of classifications, wherein the log statements comprise training data created from logs used in an enterprise computing environment including statements from a plurality of sources in the enterprise computing environment, and
the training data was created by sampling the statements in the logs and tagging the log statements in the sample with the at least one of the plurality of classifications;
identifying classified new log statements as log statements having the problem classification; and
presenting the identified classified new log statements having the problem classification through a user interface.

2. The system of claim 1 , wherein the statements contained in the logs from the plurality of sources comprise unstructured data.

3. The system of claim 1 , wherein the sampling of statements is random.

4. The system of claim 1 , wherein the instructions are further for performing the tagging of the log statements of the sample.

5. The system of claim 1 , wherein the plurality of classifications are domain specific.

6. The system of claim 1 , wherein the NLP classifier model is one of a plurality of trained models.

7. The system of claim 6 , wherein outputs from the trained NLP classifier model are provided as input to another trained model of the plurality of trained models for identifying a type of problem associated with the statements having the problem classification.

8. A method for diagnosing problems from logs, the method comprising:
receiving new log statements;
classifying the new log statements into at least one of a plurality of classifications based on an NLP feature associated with the log statements utilizing a NLP classifier model, wherein the plurality of predefined classifications include a problem classification and wherein the NLP classifier model was trained by:
utilizing log statements tagged with the at least one of the plurality of classifications, wherein the log statements comprise training data created from logs used in an enterprise computing environment including statements from a plurality of sources in the enterprise computing environment, and
the training data was created by sampling the statements in the logs and tagging the log statements in the sample with the at least one of the plurality of classifications;

identifying classified new log statements as log statements having the problem classification; and
presenting the identified classified new log statements having the problem classification through a user interface.

9. The method of claim 8 , wherein the statements contained in the logs from the plurality of sources comprise unstructured data.

10. The method of claim 8 , wherein the sampling of statements is random.

11. The method of claim 8 , further comprising performing the tagging of the log statements of the sample.

12. The method of claim 8 , wherein the plurality of classifications are domain specific.

13. The method of claim 8 , wherein the NLP classifier model is one of a plurality of trained models.

14. The method of claim 13 , wherein outputs from the trained NLP classifier model are provided as input to another trained model of the plurality of trained models for identifying a type of problem associated with the statements having the problem classification.

15. A non-transitory computer readable medium, comprising instructions for:
receiving new log statements;
classifying the new log statements into at least one of a plurality of classifications based on an NLP feature associated with the log statements utilizing a NLP classifier model, wherein the plurality of predefined classifications include a problem classification and wherein the NLP classifier model was trained by:
utilizing log statements tagged with the at least one of the plurality of classifications, wherein the log statements comprise training data created from logs used in an enterprise computing environment including statements from a plurality of sources in the enterprise computing environment, and
the training data was created by sampling the statements in the logs and tagging the log statements in the sample with the at least one of the plurality of classifications;

identifying classified new log statements as log statements having the problem classification; and
presenting the identified classified new log statements having the problem classification through a user interface.

16. The non-transitory computer readable medium of claim 15 , wherein the statements contained in the logs from the plurality of sources comprise unstructured data.

17. The non-transitory computer readable medium of claim 15 , wherein the sampling of statements is random.

18. The non-transitory computer readable medium of claim 15 , wherein the instructions are further for performing the tagging of the log statements of the sample.

19. The non-transitory computer readable medium of claim 15 , wherein the plurality of classifications are domain specific.

20. The non-transitory computer readable medium of claim 15 , wherein the NLP classifier model is one of a plurality of trained models.

21. The non-transitory computer readable medium of claim 20 , wherein outputs from the trained NLP classifier model are provided as input to another trained model of the plurality of trained models for identifying a type of problem associated with the statements having the problem classification.

Assignees

• Open Text Corp

Inventors

• Sharma Ankur
• Somayaji Ravikanth

Classifications

• G06F40/30

  Semantic analysis · CPC title

• G06F8/70

  Software maintenance or management · CPC title

• G06F40/205 Primary

  Parsing · CPC title

• G06F16/3329

  Natural language query formulation · CPC title

• G06N20/00 Primary

  Machine learning · CPC title