Computer Code Generator

Patent metadata

Field	Value
Publication number	US-2025045028-A1
Application number	US-202318365326-A
Country	US
Kind code	A1
Filing date	Aug 4, 2023
Priority date	Aug 4, 2023
Publication date	Feb 6, 2025
Grant date	—

Abstract

Disclosed herein is a system and method for generating computer code for a plurality of components of a software development project. An artificial intelligence code generator can generate computer code in response to a natural language text input describing a component of the software development project. A first database can store natural language text describing components of the software development project. A second database can store computer code generated at least partially by the artificial intelligence code generator and defining components of the software development project and corresponding to the natural language text stored in the first database. Using a pre-trained language model, the system can generate a natural language summary text based on the code intent of the first component, the identified characteristics of the objects, and the natural language text retrieved from the first database.

First claim

The invention claimed is:

1 . A system for generating computer code for a one or more components of a software development project, the system comprising:
an artificial intelligence code generator generating computer code in response to a natural language text input describing a component of the software development project; a first database storing natural language text describing components of the software development project used with the artificial intelligence code generator; a second database storing computer code generated at least partially by the artificial intelligence code generator and defining components of the software development project and corresponding to the natural language text stored in the first database; and one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to execute the steps of:
receive design requirements of the software development project;
generate a dependency graph of components of the software development project based on the design requirements;
determine from the dependency graph a subset of components of the software development project having one or more dependencies related to a first component of the software development project;
determine from attributes of the first component a code intent;
retrieve from the first database natural language text describing one or more components of the software development project having one or more dependencies related to the first component that have been used to generate computer code stored in the second database using the artificial intelligence code generator;
retrieve from the second database computer code corresponding to the retrieved natural language text describing the one more components of the software development project having one or more dependencies related to the first component;
identify from the computer code retrieved from the second database characteristics of objects within the computer code;
using a pre-trained language model, generate for the first component a natural language summary text based on the code intent of the first component, the identified characteristics of the objects, and the natural language text retrieved from the first database;
provide the natural language summary text for the first component to the artificial intelligence code generator; and
the artificial intelligence code generator generating computer code for the first component based on the provided natural language summary.

2 . The system of claim 1 , wherein the instructions further cause the one or more computers to execute the step of:
before determining the subset of components of the software development project having one or more dependencies related to the first component, receiving information describing the first component of the plurality of components of the software development project, wherein the attributes of the first component are determined from the received information.

3 . The system of claim 2 , wherein the received information is a title of the first component.

4 . The system according to claim 1 , wherein the instructions further cause the one or more computers to execute the step of:
storing the computer code for the first component in the second database and the natural language summary text for the first component in the first database.

5 . The system according to claim 1 , wherein the design requirements of the software development project comprise any one or more of: high level design requirements, system architecture, project management path, and change request proposals.

6 . A computer implemented method for generating computer code for one or more components of a software development project, the method comprising the steps of:
receiving design requirements of a software development project; generating a dependency graph of components of the software development project based on the design requirements; determining from the dependency graph a subset of components of the software development project having one or more dependencies related to a first component of the software development project; determining from attributes of the first component a code intent; retrieving from a first database storing natural language text describing components of the software development project used with an artificial intelligence code generator, natural language text describing one or more components of the software development project having one or more dependencies related to the first component that have been used to generate computer code stored in the second database using the artificial intelligence code generator; retrieving from a second database storing computer code generated at least partially by the artificial intelligence code generator and defining components of the software development project and corresponding to the natural language text stored in the first database, computer code corresponding to the retrieved natural language text describing the one more components of the software development project having one or more dependencies related to the first component; identifying from the computer code retrieved from the second database characteristics of objects within the computer code; using a pre-trained language model, generating for the first component a natural language summary text based on the code intent of the first component, the identified characteristics of the objects, and the natural language text retrieved from the first database; providing the natural language summary text for the first component to the artificial intelligence code generator; and generating computer code for the first component using the artificial intelligence code generator, based on the provided natural language summary.

7 . The method of claim 6 further comprising the step of:
before determining the subset of components of the software development project having one or more dependencies related to the first component, receiving information describing the first component of the plurality of components of the software development project, wherein the attributes of the first component are determined from the received information.

8 . The method of claim 7 , wherein the information describing the first component of the plurality of components is received from a graphical user interface, GUI.

9 . The method of claim 7 , wherein the received information is a title of the first component.

10 . The method according to claim 6 further comprising the step of:
storing the computer code for the first component in the second database and the natural language summary text for the first component in the first database.

11 . The method according to claim 6 , wherein the design requirements of the software development project comprise any one or more of: high level design requirements, system architecture, project management path, and change request proposals.

12 . The method according to claim 6 , wherein the step of determining a subset of components of the software development project having one or more dependencies related to a first component further comprises the step of flattening, tokenizing, and/or resolving direction dependencies between components in the subset of components within the dependency graph.

13 . The method according to claim 6 further comprising the step of:
before providing the natural language summary text for the first component to the artificial intelligence code generator, splitting the natural language summary text into smaller natural language summary texts each below a pre

Assignees

• Accenture Global Solutions Ltd

Inventors

• Sikand Samarth
• Singi Kapil
• Dey Kuntal
• Kaulgud Vikrant
• Sharma Vibhu Saujanya
• Burden Adam Patten
• Kiran Velama Ravi

Classifications

• G06F8/30

  Creation or generation of source code · CPC title

• G06F8/20

  Software design · CPC title

• G06F8/36

  Software reuse · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F8/10

  Requirements analysis; Specification techniques · CPC title