2023 Fiscal Year Final Research Report
The investigation of interactions among components of L2 writing ability
| Project/Area Number |
21K13065
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 02100:Foreign language education-related
|
|---|
| Research Institution | Matsuyama University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | L2ライティング / ネットワーク分析 / 学習者コーパス / CAF / 言語産出能力 / 生成AI / 自動採点 |
|---|
| Outline of Final Research Achievements |
The purpose of this study was to elucidate the representation process of English writing proficiency by examining the interactions among its components using network analysis. Network analysis was conducted based on the correlation coefficients among the observed variables that constitute linguistic and discourse knowledge. The results did not yield a statistically significant conclusion that the network changes according to differences in learners' proficiency levels. Additionally, the use of generative artificial intelligence, such as ChatGPT, demonstrated potential for the automatic calculation of accuracy indices on large-scale data.
|
| Free Research Field |
応用言語学
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究の学術的な意義として、英語ライティング能力の構成要素のうちの一つである言語能力の下位概念間のシステム（系）は熟達度に応じて必ずしも変化するというわけではないことが明らかになった。社会的意義として、予期せぬ生成AIの登場により、これまで人的及び経済的制約があり、一度に大規模な算出が困難であった正確性指標の計算が、生成AI及び自然言語処理の技術を活用することで、自動化できうることを明らかにしたことである。
|
