Automatic pre-detection of potential coding issues and recommendation for resolution actions

Patent metadata

Field	Value
Publication number	US-9519477-B2
Application number	US-201314028048-A
Country	US
Kind code	B2
Filing date	Sep 16, 2013
Priority date	Sep 16, 2013
Publication date	Dec 13, 2016
Grant date	Dec 13, 2016

Abstract

A tool for automatic pre-detection of potential software product impact according to a statement placed in a software development system, and for automatically recommending for resolutions which accesses a repository of information containing a history of changes and effects of the changes for a software project; using a received a statement in natural language to perform a natural language search of the repository; according to the findings of the search of the repository, using a machine learning model to compose an impact prediction regarding the received statement relative to the findings; and automatically placing an advisory notice regarding to the impact prediction into the software development system, wherein the advisory notice is associated with the received statement.

First claim

What is claimed is:

1. A method for automatic pre-detection of potential software product impact and recommendation for resolutions, the method comprising the steps of:
accessing by a computer processor a repository of information containing a history of software project changes and effects attributed to the software project changes subsequent to implementation;
receiving by a computer processor from a user an input phrase expressed in natural language;
performing by a computer processor natural language processing to extract symbols from the input phrase, wherein the natural language processing comprises one or more processes selected from the group consisting of deductive logic, inductive logic, validity and soundness checks, rules of thought, truth functionalities, predicate logic, sorites arguments, ethymemes, syntactic analysis, semantic analysis, and pragmatics;
searching, the repository, by a computer processor, on the extracted symbols which define an unimplemented software project change, wherein the searched symbols exclude symbols which define effects of changes;
receiving by a computer processor from the searching one or more effects attributed to previously-made software project changes which are similar to symbols which define the unimplemented software project change;
using by a computer processor, one or more machine learning models to predict one or more components of the software project likely to be impacted subsequent to implementation of the unimplemented software project change, wherein the prediction is generated from the received effects; and
placing by a computer processor an advisory notice according to the impact prediction into a software development tool, wherein the advisory notice is associated with the input phrase.

2. The method as set forth in claim 1 wherein the input phrase comprises a proposed change to the software project, and wherein the impact prediction comprises a warning of one or more software errors determined to be most probable to be induced by programming changes corresponding to the proposed change.

3. The method as set forth in claim 2 wherein the warning further comprises a suggestion for an alternative change proposition.

4. The method as set forth in claim 1 wherein the input phrase comprises a proposed requirement for the software project, and wherein the impact prediction comprises a warning of a noncompliance with another requirement determined to be most probable to be induced by programming changes corresponding to the proposed requirement.

5. The method as set forth in claim 1 wherein the input phrase comprises a description of an untested revision of the software project, and wherein the impact prediction comprises a suggestion of one or more test cases to be employed to regression test and validate the untested revision of the software project.

6. The method as set forth in claim 5 wherein the suggestion of the one or more test cases is selected from the group consisting of a recommended applicable test case, a non-recommended inapplicable test case, and a modified test suite of a plurality of recommended test cases.

7. The method as set forth in claim 1 further comprising, responsive to receiving by a computer processor a user indication of approval of the impact prediction, updating by the computer processor of the one or more machine learning models.

8. A computer program product for automatic pre-detection of potential software product impact and recommendation for resolutions comprising:
a computer-readable storage memory device excluding a propagating signal per se; and
program instructions stored by the computer readable storage memory device for causing a processor to:
access a repository of information containing a history of software project changes and effects attributed to the software project changes subsequent to implementation;
receive from a user an input phrase expressed in natural language;
perform natural language processing to extract symbols from the input phrase, wherein the natural language processing comprises one or more processes selected from the group consisting of deductive logic, inductive logic, validity and soundness checks, rules of thought, truth functionalities, predicate logic, sorites arguments, ethymemes, syntactic analysis, semantic analysis, and pragmatics;
search the repository on the extracted symbols which define an unimplemented software project change, wherein the searched symbols exclude symbols which define effects of changes;
receive from the searching one or more effects attributed to previously-made software project changes which are similar to symbols which define the unimplemented software project change;
use one or more machine learning models to predict one or more components of the software project likely to be impacted subsequent to implementation of the unimplemented software project change, wherein the prediction is generated from the received effects; and
place an advisory notice according to the impact prediction into a software development tool, wherein the advisory notice is associated with the input phrase.

9. The computer program product as set forth in claim 8 wherein the input phrase comprises a proposed change to the software product, and wherein the impact prediction comprises a warning of one or more software errors determined to be most probable to be induced by programming changes corresponding to the proposed change.

10. The computer program product as set forth in claim 9 wherein the warning further comprises a suggestion for an alternative change proposition.

11. The computer program product as set forth in claim 8 wherein the input phrase comprises a proposed requirement for the software product, and wherein the impact prediction comprises a warning of a noncompliance with another requirement determined to be most probable to be induced by programming changes corresponding to the proposed requirement.

12. The computer program product as set forth in claim 8 wherein the input phrase comprises a description of an untested revision the software project, and wherein the impact prediction comprises a suggestion of one or more test cases to be employed to regression test and validate the revision of the software project.

13. The computer program product as set forth in claim 12 wherein the suggestion of the one or more test cases is selected from the group consisting of a recommended applicable test case, a non-recommended inapplicable test case, and a modified test suite of a plurality of recommended test cases.

14. The computer program product as set forth in claim 8 further comprising sixth program instructions for, responsive to user indication of approval of the impact prediction, updating the one or more machine learning models, wherein the sixth programming instructions are stored by the computer readable storage memory device.

15. A system for automatic pre-detection of potential software product impact and recommendation for resolutions comprising:
a computing platform having a processor and a computer readable storage memory device; and
program instructions encoded by the computer readable storage memory device for causing the processor to:
access a repository of information containing a history of software project changes and effects attributed to the software project changes subsequent to implementation;
receive from a user an input phrase expressed in natural language;
perform natural language processing to extract symbols from the input phrase, wherein the natural language processing comprises one or more processes selected from a group consisting of d

Assignees

• IBM

Inventors

• Champlin-Scharff George Howard
• Snow Timothy Bruce
• Yee Sabrina

Classifications

• G06F9/451

  Execution arrangements for user interfaces · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06F8/30

  Creation or generation of source code · CPC title

• G06F9/454

  Multi-language systems; Localisation; Internationalisation · CPC title

• G06F8/70 Primary

  Software maintenance or management · CPC title