AI Text Generation Support via Morpheme Masking and Substitution

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current AI technologies are insufficient for generating sentences effectively, despite achieving high performance in tasks like document classification, translation, summarization, and Q&A.

Innovation Solution

A document generation supporting device that performs morpheme analysis, sets mask words, and uses a trained natural language processing model to search for and determine substitute possibilities for mask words, thereby supporting the generation of new documents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If AI technologies are used for document processing tasks, then performance in classification, translation, summarization, and Q&A is improved, but capability in generating sentences is insufficient

Engineering Contradiction:
Improveperformance in document processing tasksVSAvoidcapability in generating sentences
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent segments the document processing task into morpheme analysis, mask word setting, and word search components. By dividing the document into morphemes and creating masked word problems, the system enables AI to generate new sentences through controlled substitution rather than attempting complete sentence generation from scratch, thus improving sentence generation capability while maintaining document processing performance

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of word substitution likelihood by using a trained natural language processing model to evaluate and replace words based on their likelihood scores. This parameter change enables the system to generate varied sentences while maintaining contextual accuracy, addressing the sentence generation insufficiency without compromising document processing reliability

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If a trained natural language processing model is used to determine substitute possibilities, then document generation support is improved, but device complexity increases

Engineering Contradiction:
Improvedocument generation supportVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent makes the trained natural language processing model serve multiple functions: it performs morpheme analysis, evaluates word substitution likelihood, and generates substitute words. By making this single component multi-functional, the system achieves improved document generation support without proportionally increasing device complexity, as one model handles multiple tasks rather than requiring separate specialized components

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250139362A1Text generation support apparatus, text generation support method, and program
Publication Date: 2025.05.01 NIPPON TELEGRAPH & TELEPHONE CORP
  • US20250139362A1 patent drawing
  • US20250139362A1 patent drawing
  • US20250139362A1 patent drawing

AI summary

In recent years, high performance can be obtained for tasks such as document classification, translation, summarization, and Q&A, but AI alone is insufficient in terms of generating sentences, and it can be said that it is a high demand for AI to support people. An object of the present invention is to support generation of a document by generating a new sentence by supporting an idea of a person.According to the present invention, a document generation supporting device that supports generation of a document includes: a morpheme analysis unit 12 that performs morpheme analysis on data of the document and dividing the document into words; a mask word setting unit 13 that sets a mask word by masking a predetermined word among the divided words; and a word search unit 14 that uses a trained natural language processing model 2 to search for a likelihood in consideration of a word possibility and a context based on the mask word, and determines a substitute possibility for the mask word in accordance with the likelihood to complete a document possibility.