Computerized system for evaluating the likelihood of technology change incidents

• US 10,366,337 B2
• Filed: 02/24/2016
• Issued: 07/30/2019
• Est. Priority Date: 02/24/2016
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Kephart, Aaron D., Howie, Charles C., Glover, DeAundra K.
• Primary Examiner(s): Waldron, Scott A.
• Assistant Examiner(s): Tamulonis, Fen Christopher
• Application Number:
  US15/052,573
• Publication Number:
  US 20170243129A1
• Time in Patent Office: 1,252 Days
• CPC Class Codes: G06F 16/24564 Applying rules; Deductive queries
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Embodiments of the present invention relate to apparatuses, systems, methods and computer program products for a technology configuration system. Specifically, the system typically provides operational data processing of a plurality of records associated with information technology operational activities, for dynamic transformation of data and evaluation of interdependencies of technology resources. In other aspects, the system typically provides technical language processing of the plurality of records for transforming technical and descriptive data, and constructing categorical activity records. The system may be configured to achieve significant reduction in memory storage and processing requirements by performing categorical data encoding of the plurality of records. The system may employ a dynamic categorical data decoding process, which delivers a reduction in processing time when the encoded records are decoded for evaluating the exposure of technology change events to technology incidents and modifying such technology change events.

Claims

1. A computerized system for evaluating the likelihood of technology change incidents, comprising:

a computer apparatus including a processor, a memory, and a network communication device; and

a technology change evaluation module stored in the memory, executable by the processor, and configured for;

determining a decoding divisor;

retrieving a plurality of encoded records regarding a plurality of historic information technology operational activities from an activity record database based on a division operation of the decoding divisor and one or more encoded words of each encoded record of the plurality of encoded records, wherein each encoded word is associated with a technology operational activity and is divisible by the decoding divisor, wherein the division operation yields a remainder of zero;

decoding, using a categorical data decoding algorithm, each of the plurality of encoded records into a plurality of decoded records, each of the decoded records comprising a binary value in each of a plurality of data fields, the plurality of data fields including a first data field defining whether one of the historic information technology operational activities is associated with a prior technology incident;

transforming, for each decoded record of the decoded records, the binary value associated with at least one data field of the plurality of data fields into technology identifier data;

processing the decoded records using a technology incident predictive model to produce an incident predictive algorithm for predicting whether a technology change event will cause a technology incident, the incident predictive algorithm defining a subset of the data fields and a weight factor for each data field in the subset of the data fields;

retrieving a change record related to a future technology change event, the change record comprising change information related to one or more of the plurality of data fields; and

evaluating the change information in the change record using the incident predictive algorithm to determine a likelihood that the future technology change event will cause a future technology incident.

2. The computerized system according to claim 1, wherein:

the plurality of decoded records are associated with a first time period;

the technology change evaluation module is configured for incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm.

3. The computerized system according to claim 1, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field.

4. The computerized system according to claim 1, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field.

5. The computerized system according to claim 1, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.

6. The computerized system according to claim 1, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

the plurality of decoded records are associated with a first time period;

the technology change evaluation module is configured for incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises;

processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm;

performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field;

performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field; and

performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.

7. The computerized system according to claim 1, wherein the technology change evaluation module is configured for periodically updating the incident predictive algorithm.

8. A computer program product for evaluating the likelihood of technology change incidents, comprising a non-transitory computer-readable storage medium having computer-executable instructions for:

determining a decoding divisor;

retrieving a plurality of encoded records regarding a plurality of historic information technology operational activities from an activity record database based on a division operation of the decoding divisor and one or more encoded words of each encoded record of the plurality of encoded records, wherein each encoded word is associated with a technology operational activity and is divisible by the decoding divisor, wherein the division operation yields a remainder of zero;

decoding, using a categorical data decoding algorithm, each of the plurality of encoded records into a plurality of decoded records, each of the decoded records comprising a binary value in each of a plurality of data fields, the plurality of data fields including a first data field defining whether one of the historic information technology operational activities is associated with a prior technology incident;

transforming, for each decoded record of the decoded records, the binary value associated with at least one data field of the plurality of data fields into technology identifier data;

processing the decoded records using a technology incident predictive model to produce an incident predictive algorithm for predicting whether a technology change event will cause a technology incident, the incident predictive algorithm defining a subset of the data fields and a weight factor for each data field in the subset of the data fields;

retrieving a change record related to a future technology change event, the change record comprising change information related to one or more of the plurality of data fields; and

evaluating the change information in the change record using the incident predictive algorithm to determine a likelihood that the future technology change event will cause a future technology incident.

9. The computer program product according to claim 8, wherein:

the plurality of decoded records are associated with a first time period;

the non-transitory computer-readable storage medium has computer-executable instructions for incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm.

10. The computer program product according to claim 8, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field.

11. The computer program product according to claim 8, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field.

12. The computer program product according to claim 8, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.

13. The computer program product according to claim 8, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

the plurality of decoded records are associated with a first time period;

the non-transitory computer-readable storage medium has computer-executable instructions for incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises;

processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm;

performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field;

performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field; and

performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.

14. The computer program product according to claim 8, wherein the non-transitory computer-readable storage medium has computer-executable instructions for periodically updating the incident predictive algorithm.

15. A computerized method for evaluating the likelihood of technology change incidents, comprising:

determining, via a computer processor, a decoding divisor;

retrieving, via a computer processor, a plurality of encoded records regarding a plurality of historic information technology operational activities from an activity record database based on a division operation of the decoding divisor and one or more encoded words of each encoded record of the plurality of encoded records, wherein each encoded word is associated with a technology operational activity and is divisible by the decoding divisor, wherein the division operation yields a remainder of zero;

decoding, via a computer processor, using a categorical data decoding algorithm, each of the plurality of encoded records into a plurality of decoded records, each of the decoded records comprising a binary value in each of a plurality of data fields, the plurality of data fields including a first data field defining whether one of the historic information technology operational activities is associated with a prior technology incident;

transforming, via a computer processor, for each decoded record of the decoded records, the binary value associated with at least one data field of the plurality of data fields into technology identifier data;

processing, via a computer processor, the decoded records using a technology incident predictive model to produce an incident predictive algorithm for predicting whether a technology change event will cause a technology incident, the incident predictive algorithm defining a subset of the data fields and a weight factor for each data field in the subset of the data fields;

retrieving, via a computer processor, a change record related to a future technology change event, the change record comprising change information related to one or more of the plurality of data fields; and

evaluating, via a computer processor, the change information in the change record using the incident predictive algorithm to determine a likelihood that the future technology change event will cause a future technology incident.

16. The computerized method according to claim 15, wherein:

the plurality of decoded records are associated with a first time period;

the method comprises incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm.

17. The computerized method according to claim 15, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field.

18. The computerized method according to claim 15, wherein processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field.

19. The computerized method according to claim 15, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.

20. The computerized method according to claim 15, wherein:

the incident predictive algorithm defines an aggregate weight factor for the subset of the data fields;

the plurality of decoded records are associated with a first time period;

the computerized method comprises incrementally altering the first time period to identify a second time period that correlates with the first data field, the second time period being associated with a subset of the plurality of decoded records;

processing the decoded records using the technology incident predictive model to produce the incident predictive algorithm comprises;

processing the subset of the plurality of decoded records using the technology incident predictive model to produce the incident predictive algorithm;

performing a field selection test to identify the subset of the data fields, the subset of the data fields correlating with the first data field;

performing a field weight test to identify the weight factor for each data field in the subset of data fields, each weight factor correlating with the first data field; and

performing an aggregate weight test to identify the aggregate weight factor for the subset of the data fields, the aggregate weight factor correlating with the first data field.