Language Model Conversion of Software Specs to OO Model Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing software development techniques, such as those described in Japanese Unexamined Patent Application Publication No. 2004-287695, limit the development efficiency by requiring class names to be generated within a predefined vocabulary, restricting the flexibility of natural language descriptions for software specifications.

Innovation Solution

An information processing apparatus and method that utilizes a language model to convert natural language software specifications into structured model data through a series of input and acquisition processes, enabling efficient object-oriented software design by breaking down specifications into sentences and converting them into model data using predefined sentence types and object-oriented analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a predefined vocabulary is used to generate class names, then the structure and consistency of the ontology is maintained, but the flexibility and expressiveness of natural language descriptions are limited

Engineering Contradiction:

Improveflexibility of natural language descriptionsVSAvoidcomplexity of vocabulary management

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent changes the parameter of class name generation from being constrained to predefined vocabulary labels to being generated freely based on natural language input. The language model transforms class names from fixed vocabulary items to dynamically generated terms that accurately reflect the software specification, thereby improving adaptability while maintaining ontology structure through the model's training.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If manual conversion of natural language to model data is performed, then precision and control over the conversion process is maintained, but development time and labor effort increase

Engineering Contradiction:

Improvesoftware development efficiencyVSAvoidtime for specification conversion

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical process of converting natural language specifications to model data with an automated language model. The language model performs the transformation from natural language to structured model data automatically, significantly improving productivity while reducing the time and effort required for this conversion process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The language model enables the system to perform self-service by automatically converting natural language specifications into model data without requiring manual intervention. The model processes the conversion autonomously, transforming the specification directly into the required model format and improving development efficiency.

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If a basic ontology with fixed vocabulary is used, then the consistency of object-oriented design is ensured, but the ability to handle diverse and evolving software requirements is reduced

Engineering Contradiction:

Improveability to handle diverse requirementsVSAvoidconsistency of object-oriented design

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces dynamics by replacing the static fixed vocabulary with a dynamic language model that can generate class names adaptively based on the input specification. The model maintains consistency through its training on object-oriented design principles while providing the flexibility to handle diverse and evolving requirements through natural language processing.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS20250362883A1Information processing apparatus, method, and non-transitory computer-readable medium

Publication Date: 2025.11.27 NEC CORP

AI summary

An information processing apparatus comprises: at least one memory storing instructions; and at least one processor configured to execute the instructions to; input, to a predetermined language model, a first input text including a first instruction sentence for converting a required specification sentence in which a required specification of software is described in a natural language into a set of sentences in accordance with a predetermined sentence type and the required specification sentence; acquire the set of sentences converted from the required specification sentence based on the first instruction sentence; input, to the language model, a second input text including a second instruction sentence for converting the set of sentences into model data by a software design method based on object orientation and the set of sentences; and acquire model data converted from the set of sentences based on the second instruction sentence.