• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2020 Fiscal Year Final Research Report

Development of a Rule Conformance Judgment Technique using Attention Model Considering Emphasized Bias

Research Project

  • PDF
Project/Area Number 18K11452
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionOsaka Institute of Technology

Principal Investigator

Taira Hirotoshi  大阪工業大学, 情報科学部, 准教授 (20396146)

Co-Investigator(Kenkyū-buntansha) 小林 一郎  お茶の水女子大学, 基幹研究院, 教授 (60281440)
Project Period (FY) 2018-04-01 – 2021-03-31
Keywords規則適合判定技術
Outline of Final Research Achievements

In order to create a society in which robots and other devices can correctly understand and follow rules written in the language that humans use every day, we aim to establish a technology that automatically judges whether or not matters written in text conform to rules based on rules written in text. In this study, as an example, we took up the driving test for a standard car license, created large-scale evaluation data, and developed a rule conformance judgment technique using a large-scale general-purpose language expression model. We clarified that the technique can automatically learn a more accurate model than the conventional method, based on the ratio of synonyms included.

Free Research Field

自然言語処理

Academic Significance and Societal Importance of the Research Achievements

本研究では、自然言語で書かれた規則に基づき、現在の状況を自然言語で表した文が規則に合致するか否かの適合判定を高精度に行う技術を実現しようとしている。技術的には、汎用言語表現モデルによる深層学習技術を用いて、使用されている知識が一定の枠内になる限られたドメインについて、自然文で書かれた規則に基づく規則適合判定モデルの自動学習を実現するものである。本研究の目指す規則適合判定技術は、これから人間社会に普及する可能性の高い自動運転車や自律型ロボットが法令を順守しながら人間社会の中で活動するために必要不可欠な技術として社会的な意義を持つ。

URL: 

Published: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi