| Project/Area Number |
18K18118
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61030:Intelligent informatics-related
|
|---|
| Research Institution | Tokyo Gakugei University (2021)
Shizuoka Institute of Science and Technology (2018-2020) |
|---|
Principal Investigator |
Ehara Yo 東京学芸大学, 教育学部, 講師 (60738029)
|
|---|
| Project Period (FY) |
2018-04-01 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2021: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2020: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2019: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 学習支援システム / 項目反応理論 / テスト理論 / BERT / 深層学習 / 自然言語処理 / 自動作問 / 文脈化単語埋め込み / 語学学習支援 / 機械学習 / クラウドソーシング / 外国語教育 / 表現学習 |
|---|
| Outline of Final Research Achievements |
Adapting quickly to technological innovations during the research period, in this study, I developed a technology that uses deep transition learning to help non-native speakers capture the meaning of a word in consideration of text contexts, in order to help humans compose questions for second language learning. This technology is expected to be useful in various aspects of (semi-)automatic question generation tasks in the future, such as allowing non-native human question makers to easily understand the meaning of words in contexts and creating distractor options by considering the meaning of the word for creating multiple-choice questions. I have presented the current status and future prospects of these results in papers prepared for domestic conferences. For future work, I continue to develop this technology to make it more useful.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では、深層転移学習を利用して、人間が語学学習のための作問をする際に役立つように、非母語話者が文意を考慮して語義を捉えやすくする技術を開発した。この技術は、語にどのような意味があるか非母語話者が用例を通じて調べたり、語義を考慮して誤答選択肢を作成するなど、作問を（半）自動で行う際の様々な場面で人間を支援する際に有用である。具体的には、語学学習支援システムや自動作問システムの作成の支援につながる。
|
