Collaborative software event discourse decomposition

Patent metadata

Field	Value
Publication number	US-11720480-B2
Application number	US-202117216363-A
Country	US
Kind code	B2
Filing date	Mar 29, 2021
Priority date	Mar 29, 2021
Publication date	Aug 8, 2023
Grant date	Aug 8, 2023

Abstract

A ticket relating to a software event is received in a software development system. One or more public forums on software development is searched for the software event. Two or more topics on the software event are identified from one or more conversations from the one or more public forums that regard the software event. Two or more causes of the software event are determined by analyzing interrelations of the two or more topics.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving a ticket relating to a software event in a software development system;
searching one or more public forums on software development for the software event;
identifying two or more topics on the software event from one or more conversations from the one or more public forums that regard the software event;
ingesting code of a software application of the software event;
determining two or more discrete causes of the software event by analyzing interrelations of the two or more topics and the code;
identifying, by analyzing the code, that a solution has at least a threshold rate metric of addressing the software event based on a confidence score for the solution satisfying a threshold;
autonomously testing the solution in a testing environment that reflects a production environment in response to identifying that the solution has at least the threshold rate metric of addressing the software event, wherein the solution includes changing some of the code;
verifying that the solution addresses the software event within the testing environment; and
autonomously recommending the solution for the software event that accounts for all of the two or more discrete causes in order to address the software event in response to verifying that the solution addresses the software event within the testing environment.

2. The computer-implemented method of claim 1 , further comprising searching real-time messages of a plurality of conversations between a plurality of user devices for messages that relate to the software event, wherein the analyzing interrelations of the two or more topics includes identifying that some topics of some messages of the plurality of messages of some conversations of the plurality of conversations relate to other messages of the plurality of messages of other conversations of the plurality of conversations.

3. The computer-implemented method of claim 1 , wherein the solution is a first solution, further comprising:
generating, by analyzing the one or more conversations and the code, both the first solution and a second solution that both have at least a threshold rate metric of addressing the software event;
autonomously testing, in two separate trials, both the first and second solution in a testing environment that reflects a production environment in response to identifying that both the first and second solution has at least the threshold rate metric of addressing the software event, wherein both the first and second solution includes changing some of the code; and
verifying that the first solution addresses the software event within the testing environment that reflects the production environment with superior performance in comparison to how the second solution addresses the software event within the testing environment,
wherein the recommending the first solution is in response to verifying that the first solution addresses the software event within the testing environment with superior performance.

4. The computer-implemented method of claim 1 , where the solution includes changing some lines of the code, further comprising:
autonomously changing the some lines of the code per the solution; and
adding a comment within the code adjacent the some lines, wherein the comment includes information on at least one of the software event, the two or more topics, the one or more conversations, and the two or more discrete causes.

5. The computer-implemented method of claim 1 , wherein the solution includes condensing the code.

6. The computer-implemented method of claim 1 , wherein:
a neural network autonomously searches the one or more public forums for the software event and identifies the two or more topics and determines the two or more discrete causes of the software event substantially immediately upon detecting that the ticket is received in the software development system, and
the neural network autonomously provides a solution for the software event that accounts for all of the two or more discrete causes in order to address the software event substantially immediately upon determining the two or more discrete causes.

7. The computer-implemented method of claim 1 , further comprising:
identifying that the software event is of a domain of software events; and
sending the software event to a machine learning model that has been trained in the domain of software events to analyze the software event such that:
the machine learning model searches the one or more public forums for the software event;
the machine learning model identifies the two or more topics; and
the machine learning model determines the two or more discrete causes.

8. A system comprising:
a processor; and
a memory in communication with the processor, the memory containing instructions that, when executed by the processor, cause the processor to:
receive a ticket relating to a software event in a software development system;
search one or more public forums on software development for the software event;
identify two or more topics on the software event from one or more conversations from the one or more public forums that regard the software event;
ingest code of a software application of the software event;
determine two or more discrete causes of the software event by analyzing interrelations of the two or more topics and the code;
identify, by analyzing the code, that a solution has at least a threshold rate metric of addressing the software event based on a confidence score for the solution satisfying a threshold;
autonomously test the solution in a testing environment that reflects a production environment in response to identifying that the solution has at least the threshold rate metric of addressing the software event, wherein the solution includes changing some of the code;
verify that the solution addresses the software event within the testing environment; and
autonomously recommend the solution for the software event that accounts for all of the two or more discrete causes in order to address the software event in response to verifying that the solution addresses the software event within the testing environment.

9. The system of claim 8 , the memory containing further instructions that, when executed by the processor, cause the processor to search real-time messages of a plurality of conversations between a plurality of user devices for messages that relate to the software event, wherein the analyzing interrelations of the two or more topics includes identifying that some topics of some messages of the plurality of messages of some conversations of the plurality of conversations relate to other messages of the plurality of messages of other conversations of the plurality of conversations.

10. The system of claim 8 , wherein a neural network autonomously searches the one or more public forums for the software event and identifies the two or more topics and determines two or more discrete causes of the software event substantially immediately upon detecting that the ticket is received in the software development system, wherein the neural network autonomously provides a solution for the software event that accounts for all of the two or more discrete causes in order to address the software event substantially immediately upon determining the two or more discrete causes.

11. The system of claim 8 , the memory containing further instructions that, when executed by the processor, cause the processor to:
generate, by analyzing the one or more conversations and the code, both the first solution and a second solution that both have at least a threshold rate metric of addressing the software event;
a

Assignees

• IBM

Inventors

• Deluca Lisa Seacat
• Anders Kelley
• Dunne Jonathan D

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06F11/3698

  Environments for analysis, debugging or testing of software · CPC title

• G06F11/3688 Primary

  for test execution, e.g. scheduling of test suites · CPC title

• G06F8/30

  Creation or generation of source code · CPC title

• G06F8/73

  Program documentation · CPC title