| 研究課題/領域番号 |
22K00718
|
|---|
| 研究種目 |
基盤研究(C)
|
| 配分区分 | 基金 |
|---|
| 応募区分 | 一般 |
|---|
| 審査区分 |
小区分02100:外国語教育関連
|
|---|
| 研究機関 | 早稲田大学 (2023)
兵庫県立大学 (2022) |
|---|
研究代表者 |
ガイエド ジョンモリース 早稲田大学, グローバルエデュケーションセンター, 講師(テニュアトラック) (20868393)
|
|---|
| 研究分担者 |
クロス ジェフリーS 東京工業大学, 環境・社会理工学院, 教授 (90532044)
|
|---|
| 研究期間 (年度) |
2022-04-01 – 2025-03-31
|
| 研究課題ステータス |
交付 (2023年度)
|
| 配分額 *注記 |
3,640千円 (直接経費: 2,800千円、間接経費: 840千円)
2024年度: 910千円 (直接経費: 700千円、間接経費: 210千円)
2023年度: 1,170千円 (直接経費: 900千円、間接経費: 270千円)
2022年度: 1,560千円 (直接経費: 1,200千円、間接経費: 360千円)
|
|---|
| キーワード | AI writing / AI assessment / L2 writing / CALL / AI assistant / metacognition / artificial intelligence / AIED / AI in education / cognitive load |
|---|
| 研究開始時の研究の概要 |
Advancements in natural language processing have led to more sophisticated intelligent writing assistants which offer feedback to users. Yet, empirical research on these tools is lacking. This research addresses this gap in the literature by developing and assessing a novel writing assistant.
|
| 研究実績の概要 |
Research activities are progressing well, with several papers recently submitted to international conferences and journals. The projects are advancing, benefiting from collaborative efforts with other researchers working in the field of AI-assisted writing/assessment. Although the project started with the focus of developing an AI-assisted writing tool, new developments in the area of AI-based qualitative assessment have allowed the researchers to explore assessment as well. This has led to a broadening of the project's scope to include both writing production aids and automated assessment of English as a Foreign Language text.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
1: 当初の計画以上に進展している
理由
Presentations at upcoming conferences include our work on automated AI assessment of L2 writing. Overall, the momentum in this research project is strong with research surrounding AI in education benefitting from rapid technological gains and collaboration with other institutions.
|
| 今後の研究の推進方策 |
Development of localized LLMs that can perform as well as gold standard pay-per-token foundational models are being investigated.
|
