Development of an English corpus with visualized interrelationships among sentences in the framework of dependency grammar

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K00583
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 02060:Linguistics-related
• Research Institution: Meiji University
• Principal Investigator: Oya Masanori 明治大学, 国際日本学部, 専任教授 (60318748)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: 依存文法 / 日英対訳コーパス / 依存距離

Outline of Final Research Achievements

In this study, the results of syntactic analysis using a dependency grammar framework on a Japanese-English parallel corpus focused on the structural characteristics of each sentence and the general structural differences between Japanese and English. During this process, the research theme shifted to dependency distances, leading to presentations at various academic venues and papers published in several scholarly journals. Notably, it is found that the frequency distribution of dependency distances aligns with a specific distribution cross-linguistically, and that the average dependency distance in English sentences produced by learners varies according to their proficiency level. Particularly significant in the final year was the discovery of dependency relationships with much longer than average dependency distances, which exhibited significant differences between Japanese and English, paving the way for new research themes.

Free Research Field

言語学

Academic Significance and Societal Importance of the Research Achievements

日本語と英語の統語構造を依存構造の枠組みで明示し、その構造特性を依存距離という数値で表現しかつ比較対照することによって、これら2言語の構造的差異を従来より客観的に把握することが可能になる。さらに、日本人英語学習者が産出する英文の平均依存距離が学習者の熟達度に応じて変化するといった知見は今後の英語教育への応用可能性を秘めている。