2018 Fiscal Year Final Research Report
Development of a learning support system for science and engineering specialized courses taking into consideration differences in language background
| Project/Area Number |
16K01145
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational technology
|
|---|
| Research Institution | Okayama University of Science |
|---|
Principal Investigator |
|
|---|
| Research Collaborator |
KOBAYASHI nobuyuki
TAKEDA masahiro
SHIWAKU asami
TAKEMURA naoki
TAKAHASHI akiyoshi
OTANI takafumi
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Keywords | 教育支援 / 留学生支援 / 自然言語処理 / 学習システム / アルゴリズム学習 |
|---|
| Outline of Final Research Achievements |
For students with different linguistic backgrounds study in the specialized courses in the Science and Technology Department of Informatics, because the quality of the Japanese language of the technical terms is different, it is difficult to understand only by normal Japanese language learning. Therefore, learning support using a computer system is required. In this research subject, the following three were performed as learning support of the specialized course for Japanese language learners. (1) To build a learning system by estimating and visualizing learning curves by learning the terms appearing in lectures in specialized courses, (3) Extraction of keywords that are evaluated from free-form questionnaires as evaluation of lectures and lecture evaluation methods.
|
| Free Research Field |
教育工学
|
| Academic Significance and Societal Importance of the Research Achievements |
学習者のレベルを推定には学習曲線を利用するが、専門用語の学習ことを目的として、学習中の正答率から学習曲線のパラメータを更新することで、独自の単語の学習のシステムの提案を行った。学術的には目的を絞った学習曲線の更新を提案した。
また、アルゴリズムの学習として、手続きの並べ替えによりアルゴリズムの学習が行えるシステムを開発した。手続きの自動生成が課題で、講義資料のような関連するドメインにあわせた手続きの生成が必要である。これに対してニューラルネットワーク翻訳の最終層での単語生成確率のパラメータ調整により、ある程度の生成が実現可能となり、ドメイン依存の言語生成の手法を提案した。
|
