Artificial intelligence based rule generation for database change deployment

1 . A database change management system comprising:

a memory for storing a set of database changes of a plurality of database change types, a set of pre-change performance data and a set of post-change performance data;

a processor;

a non-transitory, computer-readable medium storing computer-executable code, said computer-executable code comprising instructions for:

correlating the set of pre-change performance data and the set of post-change performance data to the set of database changes and to generate a set of correlated performance data for the set of database changes;

generating database change numeric representations from the set of database changes and the set of correlated performance data, each database change numeric representation comprising a matrix having positions representing a plurality of database change field values and a piece of correlated performance data;

processing the database change numeric representations using a supervised machine learning model to identify a change pattern having a strong correlation to a database change impact corresponding to the piece of correlated performance data, the change pattern representing a learned correlation between the database change field values and a corresponding database change impact, wherein the supervised learning model is trained on historical database change data and associated performance data, the historical database change data and the associated performance data being formatted according to the database change numeric representations;

based on the change pattern, automatically generating a rule that applies to the first database change type; and

updating a database change deployment system with the rule, wherein the database change deployment system is configured to automatically enforce the rule to prevent deployment of a proposed change of the first database change type.

2 . The database change management system of claim 1 , wherein the set of correlated performance data comprises computer server performance data.

3 . The database change management system of claim 1 , wherein the set of correlated performance data comprises database server performance data.

4 . The database change management system of claim 1 , wherein the set of correlated performance data comprises a set of database change results.

5 . The database change management system of claim 1 , wherein the set of database changes include a plurality of change field values and wherein the database change numeric representations represent the plurality of change field values and wherein the change pattern represents a first change field value from the plurality of change field values.

6 . The database change management system of claim 1 , wherein the rule is executable to prevent a new change of the first database change type from being deployed.

7 . A computer program product comprising a non-transitory, computer-readable medium storing computer-executable code, said computer-executable code comprising instructions for:

receiving a set of database changes of a plurality of database change types, a set of pre-change performance data, and a set of post-change performance data;

correlating the set of pre-change performance data and the set of post-change performance data to the set of database changes and to generate a set of correlated performance data for the set of database changes;

generating database change numeric representations from the set of database changes and the set of correlated performance data, each database change numeric representation comprising a matrix having positions representing a plurality of database change field values and a piece of correlated performance data;

processing the database change numeric representations using a supervised machine learning model to identify a change pattern having a strong correlation to a database change impact corresponding to the piece of correlated performance data, the change pattern representing a learned correlation between the database change field values and a corresponding database change impact, wherein the supervised learning model is trained on historical database change data and associated performance data, the historical database change data and the associated performance data being formatted according to the database change numeric representations;

based on the change pattern, automatically generating a rule that applies to the first database change type; and

updating a database change deployment system with the rule, wherein the database change deployment system is configured to automatically enforce the rule to prevent deployment of a proposed change of the first database change type.

8 . The computer program product of claim 7 , wherein the set of correlated performance data comprises computer server performance data.

9 . The computer program product of claim 7 , wherein the set of correlated performance data comprises database server performance data.

10 . The computer program product of claim 7 , wherein the set of correlated performance data comprises a set of database change results.

11 . The computer program product of claim 7 , wherein the set of database changes include a plurality of change field values and wherein the database change numeric representations represent the plurality of change field values and wherein the change pattern represents a first change field value from the plurality of change field values.

12 . The computer program product of claim 7 , wherein the rule is executable to prevent a new change of the first database change type from being deployed.

13 . A database change management method comprising:

receiving a set of database changes of a plurality of database change types, a set of pre-change performance data, and a set of post-change performance data;

correlating the set of pre-change performance data and the set of post-change performance data to the set of database changes and to generate a set of correlated performance data for the set of database changes;

generating database change numeric representations from the set of database changes and the set of correlated performance data, each database change numeric representation comprising a matrix having positions representing a plurality of database change field values and a piece of correlated performance data;

processing the database change numeric representations using a supervised machine learning model to identify a change pattern having a strong correlation to a database change impact corresponding to the piece of correlated performance data, the change pattern representing a learned correlation between the database change field values and a corresponding database change impact, wherein the supervised learning model is trained on historical database change data and associated performance data, the historical database change data and the associated performance data being formatted according to the database change numeric representations;

based on the change pattern, automatically generating a rule that applies to the first database change type; and

updating a database change deployment system with the rule, wherein the database change deployment system is configured to automatically enforce the rule to prevent deployment of a proposed change of the first database change type.

14 . The database change management method of claim 13 , wherein the set of correlated performance data comprises computer server performance data.

15 . The database change management method of claim 13 , wherein the set of correlated performance data comprises database server performance data.

16 . The database change management method of claim 13 , wherein the set of correlated performance data comprises a set of database change results.

17 . The database change management method of claim 13 , wherein each database change numeric representation comprises a matrix having positions representing each of the plurality of database change types.