2022 Fiscal Year Final Research Report
Dynamic Linkage and Visualization of Associative Networks in Vocabulary Learning
| Project/Area Number |
18K12434
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 02090:Japanese language education-related
|
|---|
| Research Institution | Takushoku University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2023-03-31
|
| Keywords | 語彙学習 / 学習支援 / 連想情報 / 可視化 |
|---|
| Outline of Final Research Achievements |
In this study, we developed a support system for Japanese language learners to learn vocabulary according to association of words by Japanese native speakers, and evaluated it. The generated question sentences and answer choices were rated by Japanese language teachers, and about 80% of them were found to be suitable as learning support materials. We then asked Japanese language learners to use our system for a certain period of time and analyzed their learning data. As a result, we confirmed the effectiveness of learning by visualization of word associations.
|
| Free Research Field |
自然言語処理
|
| Academic Significance and Societal Importance of the Research Achievements |
海外における日本語学習の需要が高い中，自然言語処理技術が日本語学習の支援ツールや教材メディアとして応用されてきている．しかし，他の主要言語と比べると言語・教育研究の成果に基づいた辞書・コーパスや教材の開発は遅れていると考えられている．今日の日本語学習に関する辞書やコーパス，学習支援システムでは，単語の共起情報を用いたものが多く見られる．
本研究では，従来の共起情報ではなく人間が単語について連想した情報（連想情報）を語彙学習に用いている．研究成果の「学習支援の教材」と「学習の効果」に関する評価を通して，学習に連想情報がある程度有効であることを示すことができた．
|
