Determining a cause of an incident based on text analytics of documents

• US 8,996,532 B2
• Filed: 05/21/2012
• Issued: 03/31/2015
• Est. Priority Date: 05/21/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bhatt, Dhruv A., McNeil, Kristin E., Patel, Nitaben A.
• Primary Examiner(s): Fleurantin, Jean B
• Application Number:
  US13/476,399
• Publication Number:
  US 20130311479A1
• Time in Patent Office: 1,044 Days
• Field of Search: 707741-746, 707/748, 707/758
• US Class Current: 707/741
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/2272 Management thereof
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/31 Indexing; Data structures therefor; Storage structures

Abstract

According to one embodiment of the present invention, a system analyzes one or more change records based on text analytics using dictionaries and rules for the analysis in order to generate an index of analyzed data that represents the one or more change records. The change records each include a change and corresponding time frame for occurrence of the change. Information from a request is applied to the index of analyzed data to determine one or more candidate causes for the incident and the corresponding time frame for occurrence of the change. A time associated with the request is correlated with the corresponding time frame for occurrence of the change to identify the one or more candidate causes in the one or more change records as causes for the incident. Embodiments of the present invention further include a method and computer program product for determining causes of an incident.

Claims

1. A system for determining a root cause of an incident comprising:

a computer system including at least one processor configured to;

analyze one or more change records based on text analytics using a dictionary and rules for the analysis in order to generate an index of analyzed data that represents the one or more change records, wherein the change records each include a change and corresponding time frame for occurrence of the change, and wherein the text analytics includes dictionaries based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite;

apply information from a request for the root cause of the incident to the index of analyzed data to determine one or more change records with changes serving as one or more candidate causes for the incident, wherein applying the information includes;

correlating the request with the changes of the one or more change records based on a time associated with the request being within a specified time interval of the corresponding time frame of occurrence of the changes of the one or more change records; and

identifying the one or more candidate causes in the one or more change records as causes for the incident based on the correlation.

2. The system of claim 1, wherein the text analytics includes rules based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite.

3. The system of claim 1, wherein correlating includes computing a correlation score that indicates a relative likelihood of the occurrence of the change as a cause of the incident.

4. The system of claim 3, further comprising identifying the one or more candidate causes with a highest correlation score as a cause for the incident.

5. The system of claim 1, wherein the time associated with the request includes at least one of a time of the incident, a current time, and a date entered by a requester.

6. The system of claim 1, wherein the index of analyzed data includes one or more of annotations referencing the one or more change records, extensible markup language (XML) representations of the one or more change records, and metadata configured to provide additional change record information.

7. A computer program product for determining a root cause of an incident comprising:

a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising computer readable program code configured to;

analyze one or more change records based on text analytics using a dictionary and rules for the analysis in order to generate an index of analyzed data that represents the one or more change records, wherein the change records each include a change and corresponding time frame for occurrence of the change, and wherein the text analytics includes dictionaries based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite;

apply information from a request for the root cause of the incident to the index of analyzed data to determine one or more change records with changes serving as one or more candidate causes for the incident, wherein applying the information includes;

correlating the request with the changes of the one or more change records based on a time associated with the request being within a specified time interval of the corresponding time frame of occurrence of the changes of the one or more change records; and

identifying the one or more candidate causes in the one or more change records as causes for the incident based on the correlation.

8. The computer program product of claim 7, wherein the text analytics includes rules based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite.

9. The computer program product of claim 7, wherein correlating includes computing a correlation score that indicates a relative likelihood of the occurrence of the change as a cause of the incident.

10. The computer program product of claim 9, further comprising identifying the one or more candidate causes with a highest correlation score as a cause for the incident.

11. The computer program product of claim 7, wherein the time associated with the request includes at least one of a time of the incident, a current time, and a date entered by a requester.

12. The computer program product of claim 7, wherein the index of analyzed data includes one or more of annotations referencing the one or more change records, extensible markup language (XML) representations of the one or more change records, and metadata configured to provide additional change record information.

13. A computer-implemented method of determining a root cause of an incident comprising:

analyzing one or more change records based on text analytics using a dictionary and rules for the analysis in order to generate an index of analyzed data that represents the one or more change records, wherein the change records each include a change and corresponding time frame for occurrence of the change, and wherein the text analytics includes dictionaries based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite;

applying information from a request for the root cause of the incident to the index of analyzed data to determine one or more change records with changes serving as one or more candidate causes for the incident, wherein applying the information includes;

correlating the request with the changes of the one or more change records based on a time associated with the request being within a specified time interval of the corresponding time frame of occurrence of the changes of the one or more change records; and

identifying the one or more candidate causes in the one or more change records as causes for the incident based on the correlation.

14. The method of claim 13, wherein the text analytics includes rules based on one of an Unstructured Information Management Architecture (UIMA), an off-the-shelf text analytics suite, and a custom text analytics suite.

15. The method of claim 13, wherein correlating includes computing a correlation score that indicates a relative likelihood of the occurrence of the change as a cause of the incident.

16. The method of claim 15, further comprising identifying the one or more candidate causes with a highest correlation score as a cause for the incident.

17. The method of claim 13, wherein the time associated with the request includes at least one of a time of the incident, a current time, and a date entered by a requester.

18. The method of claim 13, wherein the index of analyzed data includes one or more of annotations referencing the one or more change records, extensible markup language (XML) representations of the one or more change records, and metadata configured to provide additional change record information.